{ "complete": true, "report": { "bucket": "emo_Malevolence_Malice", "label": "Intense Malevolence Malice", "definition": { "id": "emo_Malevolence_Malice", "category": "emotion", "head": "emo_Malevolence_Malice", "pole": "high", "label": "Intense Malevolence Malice", "floor": 3.0 }, "recipe_status": "direct_manual_recipe", "manual_provenance": [ { "id": "rank4/emotion_malevolence_malice/full-v2-sealed/sealed-selected", "mission": "emotion_malevolence_malice", "format_status": "CANONICAL", "heldout": true, "qualified": true, "source_cfg": 2.5, "source_model": { "label": "Humaneness 1.5 + real-speech rank 4", "repo": "laion/Humaneness-Voice-Large-1.5-RealSpeech-LoRAs", "revision": "fed33e554f094fcdc89fc0124aed2f9d42018714", "rank": 4 }, "metrics": { "ce": 6.387658805847168, "genuineness": 4.121339254379272, "wer": 0.0038924731182795698, "identity": 0.5900628405809403, "voiceclap": -0.07877223178744316, "fitness": 0.5018961556922115, "steering_v2": 0.3448919334596636, "raw_target": 0.07004110783338546, "target_cdf": 0.3448919334596636, "family_hit_all": null, "nominal_hit_all": null, "n": 100, "valid_fraction": 1, "by_arm": { "reference": { "n": 50, "ce": 6.3876588344573975, "genuineness": 4.123504486083984, "wer": 0.005204301075268817, "identity": 0.5900628405809403, "fitness": 0.5019804789928098, "steering_v2": 0.3464088711737567 }, "none": { "n": 50, "ce": 6.387658777236939, "genuineness": 4.11917402267456, "wer": 0.0025806451612903226, "identity": null, "fitness": 0.5018118323916131, "steering_v2": 0.3433749957455706 } } } } ], "selected_wording": { "id": "manual-11c936b4d8e6", "general": "Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.", "inline": [ "Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.", "Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.", "End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed." ], "cfg": 2.5, "recipe": "P05", "strategy": "local-Gemma-mutation", "timing_syntax": "canonical", "canonical_labels": true, "neutral_event": "keep", "parent": "cfg-2.5", "origin": "exact_frozen_manual_recipe" }, "selection": { "manual_no_adapter": { "config": { "scale": 0.0, "cfg": 3.0 }, "n": 20, "target_z": -0.2939941469772329, "target_raw": 0.12606616988778113, "ce": 6.160816097259522, "genuineness": 4.160595858097077, "identity": 0.5120614789426327, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.043134667724370954, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "manual_adapter": { "config": { "scale": 0.5, "cfg": 2.5 }, "n": 20, "target_z": -0.13310547676392814, "target_raw": 0.2103211209177971, "ce": 6.17868709564209, "genuineness": 4.1938589215278625, "identity": 0.550260478258133, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.004935668408870697, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "grid": [ { "config": { "scale": 0.0, "cfg": 3.0 }, "n": 20, "target_z": -0.2939941469772329, "target_raw": 0.12606616988778113, "ce": 6.160816097259522, "genuineness": 4.160595858097077, "identity": 0.5120614789426327, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.043134667724370954, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.25 }, "n": 20, "target_z": -0.4829589847362381, "target_raw": 0.027108156681060792, "ce": 6.164787411689758, "genuineness": 4.261953973770142, "identity": 0.538506331294775, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.016689815372228623, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.0 }, "n": 20, "target_z": -0.4860351579728278, "target_raw": 0.02549721151590347, "ce": 6.1933810949325565, "genuineness": 4.281409645080567, "identity": 0.5355429779738188, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.019653168693184852, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.0 }, "n": 20, "target_z": -0.48681641321911645, "target_raw": 0.025088080018758774, "ce": 6.159624743461609, "genuineness": 4.209235227108001, "identity": 0.5371464759111404, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": -0.01804967075586319, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.5 }, "n": 20, "target_z": -0.5039062491997187, "target_raw": 0.01613839268684387, "ce": 6.189012575149536, "genuineness": 4.246577620506287, "identity": 0.5510792709887028, "wer": 0.01, "valid_fraction": 0.95, "identity_delta": -0.004116875678300857, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 5.0 }, "n": 20, "target_z": 0.17724609015984913, "target_raw": 0.37284751906991004, "ce": 5.9944170951843265, "genuineness": 4.06206202507019, "identity": 0.4036278732120991, "wer": 0.005902777777777778, "valid_fraction": 1, "identity_delta": -0.15156827345490456, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 4.0 }, "n": 20, "target_z": 0.024230954556769064, "target_raw": 0.2927158184349537, "ce": 6.105813097953797, "genuineness": 4.176913595199585, "identity": 0.43922203220427036, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.11597411446273327, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.5 }, "n": 20, "target_z": -0.36884766633132154, "target_raw": 0.08686651960015297, "ce": 6.162404632568359, "genuineness": 4.204841995239258, "identity": 0.4804833360016346, "wer": 0.01152046783625731, "valid_fraction": 1, "identity_delta": -0.07471281066536903, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.5 }, "n": 20, "target_z": -0.13310547676392814, "target_raw": 0.2103211209177971, "ce": 6.17868709564209, "genuineness": 4.1938589215278625, "identity": 0.550260478258133, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.004935668408870697, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 3.0 }, "n": 20, "target_z": -0.24279784948121283, "target_raw": 0.1528768926858902, "ce": 6.140959405899048, "genuineness": 4.144278156757355, "identity": 0.5076640829443931, "wer": 0.008204334365325076, "valid_fraction": 1, "identity_delta": -0.04753206372261047, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.0 }, "n": 20, "target_z": -0.33427734469477655, "target_raw": 0.10497047156095504, "ce": 6.189806818962097, "genuineness": 4.190407121181488, "identity": 0.5125249765813351, "wer": 0.00804093567251462, "valid_fraction": 1, "identity_delta": -0.042671170085668564, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.5 }, "n": 20, "target_z": -0.4238769513632696, "target_raw": 0.058048519492149356, "ce": 6.1653831720352175, "genuineness": 4.240301609039307, "identity": 0.509691596031189, "wer": 0.0027777777777777775, "valid_fraction": 1, "identity_delta": -0.04550455063581467, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.0 }, "n": 20, "target_z": -0.43183593894495237, "target_raw": 0.05388051867485046, "ce": 6.160816073417664, "genuineness": 4.236535954475403, "identity": 0.5133953064680099, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.04180084019899368, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.0 }, "n": 20, "target_z": -0.4716308569352535, "target_raw": 0.033040525019168855, "ce": 6.185239839553833, "genuineness": 4.219904494285584, "identity": 0.506058732420206, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": -0.04913741424679756, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.0 }, "n": 20, "target_z": -0.47404785161529633, "target_raw": 0.03177478164434433, "ce": 6.214429163932801, "genuineness": 4.300865411758423, "identity": 0.5554310798645019, "wer": 0.01803921568627451, "valid_fraction": 0.95, "identity_delta": 0.00023493319749832154, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.0 }, "n": 20, "target_z": -0.47534179481316413, "target_raw": 0.031097163259983063, "ce": 6.168163084983826, "genuineness": 4.248460483551026, "identity": 0.5514078613370657, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.003788285329937935, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.5 }, "n": 20, "target_z": -0.506787109691778, "target_raw": 0.014629729837179185, "ce": 6.156646180152893, "genuineness": 4.262581551074982, "identity": 0.5284735500812531, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.02672259658575058, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.25 }, "n": 20, "target_z": -0.5169921902787545, "target_raw": 0.0092854842543602, "ce": 6.184644150733948, "genuineness": 4.283292508125305, "identity": 0.5295075014233589, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.025688645243644715, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.25 }, "n": 20, "target_z": -0.5216796890338725, "target_raw": 0.006830712407827377, "ce": 6.199338054656982, "genuineness": 4.226494312286377, "identity": 0.506308839470148, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.048887307196855544, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 5.0 }, "n": 20, "target_z": 0.6607482987370892, "target_raw": 0.6260502763092518, "ce": 6.024599266052246, "genuineness": 4.059551632404327, "identity": 0.4640594049822539, "wer": 0.011165935672514619, "valid_fraction": 1, "identity_delta": -0.09113674168474972, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 5.0 }, "n": 20, "target_z": 0.43564452110198837, "target_raw": 0.5081668496131897, "ce": 6.0825807571411135, "genuineness": 4.071476149559021, "identity": 0.4059830538928509, "wer": 0.01111111111111111, "valid_fraction": 0.95, "identity_delta": -0.14921309277415276, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 5.0 }, "n": 20, "target_z": 0.11000976072883446, "target_raw": 0.3376368753612041, "ce": 6.072255325317383, "genuineness": 4.118860220909118, "identity": 0.3745264193043113, "wer": 0.006666666666666666, "valid_fraction": 1, "identity_delta": -0.18066972736269235, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 4.0 }, "n": 20, "target_z": 0.02021483847036634, "target_raw": 0.2906126394867897, "ce": 6.044455981254577, "genuineness": 4.106308162212372, "identity": 0.4667731270194054, "wer": 0.003125, "valid_fraction": 1, "identity_delta": -0.08842301964759827, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 4.0 }, "n": 20, "target_z": -0.07817994233535659, "target_raw": 0.2390847884118557, "ce": 6.119315648078919, "genuineness": 4.161851108074188, "identity": 0.4421659611165524, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.11303018555045127, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 4.0 }, "n": 20, "target_z": -0.11794432728167956, "target_raw": 0.21826078444719316, "ce": 6.075630950927734, "genuineness": 4.095011281967163, "identity": 0.46878850385546683, "wer": 0.005555555555555555, "valid_fraction": 1, "identity_delta": -0.08640764281153679, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.5 }, "n": 20, "target_z": -0.15488282430414935, "target_raw": 0.19891665503382683, "ce": 6.136392378807068, "genuineness": 4.156516480445862, "identity": 0.4724458336830139, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": -0.08275031298398972, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 3.0 }, "n": 20, "target_z": -0.26733399220553034, "target_raw": 0.1400276876986027, "ce": 6.134208130836487, "genuineness": 4.202017760276794, "identity": 0.49682843387126924, "wer": 0.0027777777777777775, "valid_fraction": 1, "identity_delta": -0.058367712795734404, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 3.0 }, "n": 20, "target_z": -0.3234375080028131, "target_raw": 0.11064712926745415, "ce": 6.105813097953797, "genuineness": 4.131726086139679, "identity": 0.48851205706596373, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.06668408960103989, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.0 }, "n": 20, "target_z": -0.3638671835208235, "target_raw": 0.08947472274303436, "ce": 6.154462003707886, "genuineness": 4.242184412479401, "identity": 0.5016035363078117, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.0535926103591919, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.5 }, "n": 20, "target_z": -0.3879638700801835, "target_raw": 0.07685565426945687, "ce": 6.138378000259399, "genuineness": 4.215197479724884, "identity": 0.4788142845034599, "wer": 0.0029411764705882353, "valid_fraction": 1, "identity_delta": -0.0763818621635437, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.25 }, "n": 20, "target_z": -0.4598144522524711, "target_raw": 0.03922859653830528, "ce": 6.172134423255921, "genuineness": 4.253167498111725, "identity": 0.5026209652423859, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.05257518142461777, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.5 }, "n": 20, "target_z": -0.49667968192026085, "target_raw": 0.019922836124897002, "ce": 6.19377818107605, "genuineness": 4.225239145755768, "identity": 0.4902041494846344, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.06499199718236923, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" } ], "development_standard": [ { "scale": 0.0, "n": 20, "target_z": -0.49239502272542834, "target_raw": 0.022166647017002106, "ce": 6.219591879844666, "genuineness": 4.355153119564056, "identity": 0.5551961466670037, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.0, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 0.5, "n": 20, "target_z": -0.48984374724347546, "target_raw": 0.023502711206674576, "ce": 6.214429140090942, "genuineness": 4.325655698776245, "identity": 0.5381116345524788, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.01708451211452484, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 0.75, "n": 20, "target_z": -0.5229492168937919, "target_raw": 0.006165879964828491, "ce": 6.20291223526001, "genuineness": 4.325969636440277, "identity": 0.5708803609013557, "wer": 0.0027777777777777775, "valid_fraction": 1, "identity_delta": 0.015684214234352113, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 1.0, "n": 20, "target_z": -0.4371093759336615, "target_raw": 0.05111889988183975, "ce": 6.2219746828079225, "genuineness": 4.284861493110657, "identity": 0.5392279334366321, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.015968213230371474, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" } ] }, "heldout": { "standard": { "n": 50, "target_z": -0.49449218430332076, "target_raw": 0.021068395376205445, "ce": 6.298542051315308, "genuineness": 4.322454962730408, "identity": 0.5935348200798035, "wer": 0.00831545266327875, "valid_fraction": 1, "identity_delta": 0.0, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "manual": { "n": 50, "target_z": -0.3413183519912934, "target_raw": 0.10128320306539536, "ce": 6.209345846176148, "genuineness": 4.218021655082703, "identity": 0.5181017422676086, "wer": 0.008536231884057971, "valid_fraction": 1, "identity_delta": -0.07543307781219483, "identity_pairs": 50, "gates_pass": false, "gate_scope": "heldout_50_reference_pairs" }, "adapter": { "n": 50, "target_z": -0.4451220672997743, "target_raw": 0.046922775208950045, "ce": 6.253427677154541, "genuineness": 4.219778957366944, "identity": 0.5478835234045982, "wer": 0.012042160737812911, "valid_fraction": 1, "identity_delta": -0.04565129667520523, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "matched_no_adapter": { "n": 50, "target_z": -0.3825976525309149, "target_raw": 0.07966586083173752, "ce": 6.244373044967651, "genuineness": 4.213000855445862, "identity": 0.5576047718524932, "wer": 0.00658008658008658, "valid_fraction": 1, "identity_delta": -0.035930048227310184, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" } }, "adapter_vs_matched_cfg": { "n": 50, "delta_z": -0.06252441476885941, "ci95_z": [ -0.18345703828136056, 0.012543947608307024 ], "language_deltas": { "en": -0.009375003201125259, "de": -0.2934570422649831, "fr": -0.027539061410728212, "nl": 0.036254883209861914, "es": -0.018505850177322414 }, "languages_improved": 1, "clear_gain": false, "delta_ce": 0.009054632186889648, "delta_genuineness": 0.0067781019210815426, "delta_wer": 0.005462074157726332, "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": -0.10380371530848091, "ci95_z": [ -0.21363770182312192, -0.04147461072325172 ], "language_deltas": { "en": -0.06547852184385268, "de": -0.32099610367571124, "fr": -0.05771484842942268, "nl": -0.0397216762446231, "es": -0.03510742634879485 }, "languages_improved": 0, "clear_gain": false, "delta_ce": 0.044081830978393556, "delta_genuineness": 0.0017573022842407226, "delta_wer": 0.0035059288537549406, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "manual_vs_standard": { "n": 50, "delta_z": 0.15317383231202736, "ci95_z": [ 0.04685546950804424, 0.34452149010701494 ], "language_deltas": { "en": 0.0518066411085033, "de": 0.5305664153531999, "fr": 0.027050783106208048, "nl": 0.07910156336697143, "es": 0.07734375862525415 }, "languages_improved": 5, "clear_gain": true, "delta_ce": -0.08919620513916016, "delta_genuineness": -0.10443330764770507, "delta_wer": 0.00022077922077922058, "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": 1793036277.970411 }, "rows": [ { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "001c6fd88a00b33c5d4f597eef6d9e0bfd1a8371de97a23a9456e514af0d6144", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "6bcf4de16713a4f147c21a916fa5928e7608468038efb869b0e73ea0c1688318", "generation_seconds_batch": 26.80087589903269, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1286488026380539, "target_z": -0.28906249199718687, "ce": 6.038498878479004, "genuineness": 3.8420867919921875, "identity": 0.6461635231971741, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/001c6fd88a00b33c5d4f597eef6d9e0bfd1a8371de97a23a9456e514af0d6144.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": "00b300a8a9c2ef010bcdc2f971be452fa83eaabd01aac9707c66852d1c91e14a", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2e82e9ff1828ac096d1caf09fcd404620dcb2458ed40526d1116f58836848654", "generation_seconds_batch": 17.308502736967057, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02125251293182373, "target_z": -0.49414060343686467, "ce": 6.292664527893066, "genuineness": 4.218649387359619, "identity": 0.7012151479721069, "hypothesis": "De weg gaat linksaf na de brug. Loop verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/00b300a8a9c2ef010bcdc2f971be452fa83eaabd01aac9707c66852d1c91e14a.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "0133d95baf037b5cec33ddd4e3a3f91eb8cd7632513ac611211e6109fcbc6b62", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d29ed79fe56933a2ebdc4a2bbef4ace09bae9aaa0bd1e496862355a61310a8c0", "generation_seconds_batch": 25.118259939947166, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.014092743396759033, "target_z": -0.5078125097812161, "ce": 6.038498878479004, "genuineness": 4.406930446624756, "identity": 0.7026807069778442, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/0133d95baf037b5cec33ddd4e3a3f91eb8cd7632513ac611211e6109fcbc6b62.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "01b7df8c8d233421e3ff485b50c347264ffe93c4588e5ce06202524b601333fc", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "a286b9f2f35900bdfd26ff63997276a1c25c84621c4d7092d516941e22ac6d87", "generation_seconds_batch": 26.705004937946796, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.012047111988067627, "target_z": -0.5117187362173774, "ce": 5.796247482299805, "genuineness": 4.18099308013916, "identity": 0.7583136558532715, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/01b7df8c8d233421e3ff485b50c347264ffe93c4588e5ce06202524b601333fc.mp3" }, { "id": "heldout-fr-4", "language": "fr", "sentences": [ "J’ai écrit le numéro sur l’enveloppe.", "Lis-le bien... et vérifie les deux derniers chiffres.", "Appelle-moi quand tu auras un moment." ], "text": "J’ai écrit le numéro sur l’enveloppe. Lis-le bien... et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-4", "key": "01f5d5f22d9b51f4e674eec92be80d0710e62f5704093e3409bc811236eec3e5", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Intense Malevolence Malice. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "fb25234e1ed8ce7f7d661439803be88de850f3b97844f958140daf20d00924c5", "generation_seconds_batch": 19.97884456103202, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06727948784828186, "target_z": -0.40624999644319415, "ce": 6.0146708488464355, "genuineness": 4.300237655639648, "identity": 0.5484181046485901, "hypothesis": "J'écris le numéro sur l'enveloppe, lis le bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.09090909090909091, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/01f5d5f22d9b51f4e674eec92be80d0710e62f5704093e3409bc811236eec3e5.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "028987c5e20edb8176c6eb17d0d2d18bc554967896449e25f27317143485f4eb", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "f62398f683c8bd5647d22d356055f77c3fc104aad4d35aef0bda818577984e72", "generation_seconds_batch": 26.705004937946796, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.018184036016464233, "target_z": -0.5, "ce": 6.078212261199951, "genuineness": 4.005263805389404, "identity": 0.7084732055664062, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/028987c5e20edb8176c6eb17d0d2d18bc554967896449e25f27317143485f4eb.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "028ee67c3b8ff08a6c711002f5fdc6d8c908b3096dd4e2bd9896198038854259", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": 127, "seconds_generated": 10.16, "mp3_sha256": "a6ed98019bc6c759e070ae2397cc60c494031830345a4a68d482db264d3012dd", "generation_seconds_batch": 27.50013582292013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07750770449638367, "target_z": -0.38671875044460075, "ce": 6.324435234069824, "genuineness": 4.268857479095459, "identity": 0.5810846090316772, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/028ee67c3b8ff08a6c711002f5fdc6d8c908b3096dd4e2bd9896198038854259.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "02bf1fc980c275cb7b8218323128da3cce58ef5fb71849494ff5201d3c9a4a37", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "bebbc7ac07410ead4e084c97642be45bddb0c22edb40698ee6d9d2ed57fa60ff", "generation_seconds_batch": 24.73673979495652, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.016138404607772827, "target_z": -0.5039062264361613, "ce": 6.125868320465088, "genuineness": 4.532451152801514, "identity": 0.5665967464447021, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/02bf1fc980c275cb7b8218323128da3cce58ef5fb71849494ff5201d3c9a4a37.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "03d280feaf4dcd4571fae7a86f09aaa0372a761919139f3f89a2cc0a7dfa5f73", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": 127, "seconds_generated": 10.16, "mp3_sha256": "13f5b5acaa3ff7d4a8c823a77d861d14df68b2e7969501898f6957a301e60013", "generation_seconds_batch": 26.45458048803266, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.016591906547546387, "target_z": -0.5664062477769963, "ce": 6.046441555023193, "genuineness": 4.15588903427124, "identity": 0.6763809323310852, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/03d280feaf4dcd4571fae7a86f09aaa0372a761919139f3f89a2cc0a7dfa5f73.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "040d4eba2f6b8ba677171b7e066e0bff27dfcd36168c2cad58e88ceb8dba3f08", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "68e3b68b0a08b9a0cbcee76368d29cec3524181a3d47970ab4141bcdf7062c0b", "generation_seconds_batch": 28.450004700105637, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03659482300281525, "target_z": -0.46484376289342116, "ce": 6.038498878479004, "genuineness": 4.444586753845215, "identity": 0.5250688791275024, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/040d4eba2f6b8ba677171b7e066e0bff27dfcd36168c2cad58e88ceb8dba3f08.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "046fcb827b814d6dc36cfb64855bcd1bfae54a04291aa9a770618bd9f007bd6c", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "4a4360ec65171c84e01f159c5c4a955bccd0a2c4627128f9358d52a7cacc6ef6", "generation_seconds_batch": 26.423480157973245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02125251293182373, "target_z": -0.49414060343686467, "ce": 6.1020402908325195, "genuineness": 4.118232727050781, "identity": 0.5044209957122803, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/046fcb827b814d6dc36cfb64855bcd1bfae54a04291aa9a770618bd9f007bd6c.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": "04e7332297a5643779bc821339a158b39b65e7e7a1d239dad4e19c86e311b83e", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e37ce40690bb4a28b8b50d9597cab2524084e401ecc21e7c24a275c4b3ecab6c", "generation_seconds_batch": 18.721526676090434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.039663299918174744, "target_z": -0.4589843663302858, "ce": 6.213237762451172, "genuineness": 4.557555198669434, "identity": 0.5562314987182617, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/04e7332297a5643779bc821339a158b39b65e7e7a1d239dad4e19c86e311b83e.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": "053da8539d90ba07d33565fb52e52125a4651b04c6829cdb3a5e9a4f39281c33", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ed66092d56ba40edd479e9b286f58d09435941493d5e3be031b2ea21097b2db4", "generation_seconds_batch": 23.27837068203371, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.018184036016464233, "target_z": -0.5, "ce": 5.982900142669678, "genuineness": 3.9299514293670654, "identity": 0.6397233009338379, "hypothesis": "De weg gaat linksaf naar de brug. Loop verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.047619047619047616, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/053da8539d90ba07d33565fb52e52125a4651b04c6829cdb3a5e9a4f39281c33.mp3" }, { "id": "heldout-de-0", "language": "de", "sentences": [ "Deine Nachricht kam, als ich schon weg war.", "Ich las sie zweimal... dann legte ich das Handy weg.", "Sag mir, wann wir reden können." ], "text": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal... dann legte ich das Handy weg. Sag mir, wann wir reden können.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-0", "key": "059867605942d1f4c46c2c44d936e08b7fbe89dd4feef4c05500a7ed426c27a3", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Intense Malevolence Malice. 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": "23f72759feae9287b09a1b6e8dd828458bcd2953d932c0578cce0394e55ea5b2", "generation_seconds_batch": 18.421020167996176, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.004318058490753174, "target_z": -0.5429687753422415, "ce": 6.427689552307129, "genuineness": 4.469690799713135, "identity": 0.5804910659790039, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/059867605942d1f4c46c2c44d936e08b7fbe89dd4feef4c05500a7ed426c27a3.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": "05a1a29c793d96e8a20e6f33924917dc5b3233669befbedfe436d182444fbbef", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The keys are on the kitchen table.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "bcd865e0ce0e8214d2affa28f9b21ff9e0ac8e3e200725ce64db047a64f0d346", "generation_seconds_batch": 26.662643237039447, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.024774491786956787, "target_z": -0.5820312673394284, "ce": 6.372091293334961, "genuineness": 4.068024158477783, "identity": 0.42875993251800537, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you leave.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/05a1a29c793d96e8a20e6f33924917dc5b3233669befbedfe436d182444fbbef.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "0606a4ae4f71e41f4ec3dafe98926a77d627cebf428c5eb41da3d0d2c61e2604", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "d4b196da95e42c496397e61ddab5cda59e020711c6f16047e786ff2fd92b98ed", "generation_seconds_batch": 26.164028721046634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.043185263872146606, "target_z": -0.6171874759915607, "ce": 6.268836498260498, "genuineness": 4.413206577301025, "identity": 0.3182138502597809, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/0606a4ae4f71e41f4ec3dafe98926a77d627cebf428c5eb41da3d0d2c61e2604.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "0635be406134200202333f50a0dde8ab4ea49f7c26ab4f9ffb087baee725a501", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "6aff4f45fc29cbf9ca4f2f510069116f814abd4808627b38ffbc4cb1c1d71281", "generation_seconds_batch": 26.067440679995343, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11944340169429779, "target_z": -0.30664062477769966, "ce": 5.824047088623047, "genuineness": 4.268857479095459, "identity": 0.536397397518158, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/0635be406134200202333f50a0dde8ab4ea49f7c26ab4f9ffb087baee725a501.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": "06365ca4be80953889dff6799cdd39793b2cbc532feb2adbeca926460b88cc25", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9108aeb56acc090263547d890a6e755a663320c8cc9b00a5ed3dbb39749de398", "generation_seconds_batch": 17.156120534986258, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0059101879596710205, "target_z": -0.5234374724347548, "ce": 6.157639026641846, "genuineness": 4.030367851257324, "identity": 0.5953044891357422, "hypothesis": "Ik schreef het nummer op de envelop. Lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/06365ca4be80953889dff6799cdd39793b2cbc532feb2adbeca926460b88cc25.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "063f57c7c8fc0c9cf0736cd3470f5d3f920a483a9cf64a4a480edd17fbbb6aef", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "c9689bfaf0be7ad8761b17b22f5424fdbdbd40b9151d4cebdf4438b110bf848a", "generation_seconds_batch": 25.659666858031414, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02022966742515564, "target_z": -0.49609377356383866, "ce": 6.260893821716309, "genuineness": 4.243753433227539, "identity": 0.5430675148963928, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/063f57c7c8fc0c9cf0736cd3470f5d3f920a483a9cf64a4a480edd17fbbb6aef.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "0709690233aac4544ac4ee79f95ad04673be8a90b29e7e8d161b807a96ef1127", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "c27c136fd53234ccacaeb6310e9f70c90084a2b8e185d1af06d544712715441b", "generation_seconds_batch": 28.15937693894375, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04375457763671875, "target_z": -0.4511718850035164, "ce": 6.117925643920898, "genuineness": 4.545003414154053, "identity": 0.6357801556587219, "hypothesis": "La porte t'est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/0709690233aac4544ac4ee79f95ad04673be8a90b29e7e8d161b807a96ef1127.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "0728d2c1ccb824d58486d5a7facb0a458eb8487a492b81a65a996b010e604e2b", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": 127, "seconds_generated": 10.16, "mp3_sha256": "8bf8fbc08b838e9ea32c32afe446d8fde0284f4c50193235aa434e27fdc90878", "generation_seconds_batch": 19.430608466966078, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.37821733951568604, "target_z": 0.18750000711361167, "ce": 6.197352409362793, "genuineness": 3.860914945602417, "identity": 0.3964282274246216, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/0728d2c1ccb824d58486d5a7facb0a458eb8487a492b81a65a996b010e604e2b.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "082ae1ec438612c0791e927e23abaca0f242b84c08dcbdc1eebebdf62a59b67b", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "3bfd53da8d30463e1440abe2b2e6a50175e69aa09cc95b9211a24996929c98cf", "generation_seconds_batch": 22.91845166904386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.010001450777053833, "target_z": -0.5156250195624321, "ce": 6.213237762451172, "genuineness": 4.7081804275512695, "identity": 0.7306002378463745, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/082ae1ec438612c0791e927e23abaca0f242b84c08dcbdc1eebebdf62a59b67b.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "0867a8b404d644cee611f2eb567d0a419731e76d91f69d70ce725d3d71e7aead", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "89a1263a6351146651c49b2bbb13de36b78f5812e5de5150957e7c1afcc7383c", "generation_seconds_batch": 25.78369962598663, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03659482300281525, "target_z": -0.46484376289342116, "ce": 6.23706579208374, "genuineness": 4.369274139404297, "identity": 0.43321606516838074, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/0867a8b404d644cee611f2eb567d0a419731e76d91f69d70ce725d3d71e7aead.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": "08778f7f453a8c6a7f1c44fc6dd4635c8a272d8f7d1d8c8132010710a0b241ff", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Someone moved the little blue chair.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8b3af45fa6bedb8a0f43eb8bb14c24d59f32fb3fbec33e906b34bfd1dd231fc6", "generation_seconds_batch": 21.012329844990745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.023298144340515137, "target_z": -0.49023437700070327, "ce": 6.467403411865234, "genuineness": 4.482243061065674, "identity": 0.3858948349952698, "hypothesis": "Someone moved the little blue chair. It used to stand here beside the tall bookshelf. We can put it back after lunch.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/08778f7f453a8c6a7f1c44fc6dd4635c8a272d8f7d1d8c8132010710a0b241ff.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "094e53394708499ae89807aabbaa7352cdc03657dac7ced212444a4a6b97d6eb", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "02ee5126d0ce7c5e9dd3fb73d5168312eda08fe5e6d7199acb0ad47e05a036f6", "generation_seconds_batch": 23.85318032791838, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03148071467876434, "target_z": -0.47460938589271784, "ce": 6.181467056274414, "genuineness": 4.231201171875, "identity": 0.5082440376281738, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/094e53394708499ae89807aabbaa7352cdc03657dac7ced212444a4a6b97d6eb.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "0a3a76e396a72f13b85e8df9d311e1ebbd2532e6e50df16698ecd441c5483c81", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "8f6b5b0afda8c24a804c8463ce49eba8c9db564bf8a33d71c6499ab4bc18cda6", "generation_seconds_batch": 27.164152734098025, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4958418607711792, "target_z": 0.41210939300632954, "ce": 5.720792293548584, "genuineness": 4.080576419830322, "identity": 0.4561217427253723, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/0a3a76e396a72f13b85e8df9d311e1ebbd2532e6e50df16698ecd441c5483c81.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": "0a50be954ae58441dbd8530d84687d4f09f87c297cc7a62a41b9619bebde6744", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7707ed0e61d6faf8f19d9e1f0b281fc3850ad5499b7c1706b8601fad039b0a93", "generation_seconds_batch": 17.22793620498851, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.024774491786956787, "target_z": -0.5820312673394284, "ce": 6.260893821716309, "genuineness": 4.406930446624756, "identity": 0.6552256941795349, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai regardé dans le tiroir, puis près de la fenêtre. Prends les quand tu partiras.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/0a50be954ae58441dbd8530d84687d4f09f87c297cc7a62a41b9619bebde6744.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "0abf995497820c6b1cdcc4f8c80db09c484f291b8dc03af55fa8153ab66465da", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "74c173dcecf8300bd4916800475146b02d2fa8142d0141c0cfc0ee2d471bee75", "generation_seconds_batch": 25.586506217019632, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.16751602292060852, "target_z": -0.21484376289342114, "ce": 6.24500846862793, "genuineness": 3.848362922668457, "identity": 0.3042999505996704, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/0abf995497820c6b1cdcc4f8c80db09c484f291b8dc03af55fa8153ab66465da.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "0adbb282be63e7414ce7ea3b92ae7e37af144f406d90cea85d7fad01ef831a3a", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "be4f50312a04e14087c4a5fdc953f0120fd7c6edcc929b20d0c35f4fb68e60c0", "generation_seconds_batch": 26.544314936967567, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.016591906547546387, "target_z": -0.5664062477769963, "ce": 6.3641486167907715, "genuineness": 4.193545341491699, "identity": 0.12913736701011658, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/0adbb282be63e7414ce7ea3b92ae7e37af144f406d90cea85d7fad01ef831a3a.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": "0b0e1b2eb00031f87f4db4257b436ed908c97dabbd4b9f52440be2c5855f3ee5", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Your message arrived after I had left.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9338221389162ddc5a8efb512b1c455f33d2aefffbd2a0e58d724a2b705e9387", "generation_seconds_batch": 27.025717504904605, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04375457763671875, "target_z": -0.4511718850035164, "ce": 6.483288764953613, "genuineness": 4.15588903427124, "identity": 0.3024730682373047, "hypothesis": "Your message arrived after I had left. I read it twice before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/0b0e1b2eb00031f87f4db4257b436ed908c97dabbd4b9f52440be2c5855f3ee5.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "0b130164ebc0b6bf8ec72de7ce4b7240ee1dce7812f6ab047026b186d1736542", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "61fd48b6ab49dbe419e25509f69fa1737e5d807ba14eada0152f9c0e0a01de53", "generation_seconds_batch": 26.60191045003012, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.008409321308135986, "target_z": -0.5507812282145643, "ce": 5.974957466125488, "genuineness": 4.124508857727051, "identity": 0.483108788728714, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/0b130164ebc0b6bf8ec72de7ce4b7240ee1dce7812f6ab047026b186d1736542.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "0b3e1eb62b2ab5da0f8a2c9cf5afe63f079c202e8d69cf51404b9b1ba31341e0", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "ac7b3f5d37b6ab73f0cbba4185fb92108d873979180b38a292308b79d64d6f7f", "generation_seconds_batch": 21.796908541000448, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.030911415815353394, "target_z": -0.5937500035568058, "ce": 6.09409761428833, "genuineness": 4.570107460021973, "identity": 0.5578603744506836, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/0b3e1eb62b2ab5da0f8a2c9cf5afe63f079c202e8d69cf51404b9b1ba31341e0.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "0b479dd3f55557d653feb310c093ddb1c22db260a07e585fdcff71b0db1fdf28", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "9dc8bdcf5f0f78708d466e1ea6bba458d0ff8f2e8a218ed21f81608b0418a0d0", "generation_seconds_batch": 26.803152821026742, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.018637537956237793, "target_z": -0.5703124742131577, "ce": 6.125868320465088, "genuineness": 4.293961524963379, "identity": 0.7392690181732178, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/0b479dd3f55557d653feb310c093ddb1c22db260a07e585fdcff71b0db1fdf28.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "0bcea7d26f162e143fbfc40b3d2a779d7daf3d09f0a61c63d4f541e3c57673f6", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cd81294c9ffdff9a24590fb3e680651075a3c7d562684b8111193320825ad459", "generation_seconds_batch": 26.77258615894243, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.010001450777053833, "target_z": -0.5156250195624321, "ce": 6.300607204437256, "genuineness": 4.093128204345703, "identity": 0.1726876199245453, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/0bcea7d26f162e143fbfc40b3d2a779d7daf3d09f0a61c63d4f541e3c57673f6.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "0c2efcd5f4d6b4596e3e9619756538eac1c31114ba2765f5d6110f9b1e23c168", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "b4dd639f939c4478aa8a420049e44def3ba534203ecee97e2380642ac00fc772", "generation_seconds_batch": 26.94950179895386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06727948784828186, "target_z": -0.40624999644319415, "ce": 6.197352409362793, "genuineness": 4.099404335021973, "identity": 0.11920332163572311, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/0c2efcd5f4d6b4596e3e9619756538eac1c31114ba2765f5d6110f9b1e23c168.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "0c405ba89bdbc110d2738434b43093f7e6d30d709cd0c97dabedf8408168bf44", "bucket": "emo_Malevolence_Malice", "phase": "wording", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "2c76b8902b32b537fd45c8e0ddafab7865808061f5f2a347512ee2cb3de60fb7", "generation_seconds_batch": 19.743997538927943, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00636368989944458, "target_z": -0.5468750017784029, "ce": 6.221180438995361, "genuineness": 4.1998209953308105, "identity": 0.11957155913114548, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/0c405ba89bdbc110d2738434b43093f7e6d30d709cd0c97dabedf8408168bf44.mp3" }, { "id": "heldout-nl-4", "language": "nl", "sentences": [ "Ik schreef het nummer op de envelop.", "Lees het goed... en controleer de laatste twee cijfers.", "Bel me wanneer je even tijd hebt." ], "text": "Ik schreef het nummer op de envelop. Lees het goed... en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-4", "key": "0cb70e86df8754d3bba44fe7f798aac7d0ffe2ca68fc75b92ff1a1ed29986ba7", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Intense Malevolence Malice. 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": "771c09001fcb745fd691bc22c50555bddb8392315fa531c8056cec0e9f8a5b41", "generation_seconds_batch": 17.005645935074426, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.016138404607772827, "target_z": -0.5039062264361613, "ce": 6.268836498260498, "genuineness": 4.287685871124268, "identity": 0.7999069690704346, "hypothesis": "Ik schreef het nummer op de encelop. Lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.043478260869565216, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/0cb70e86df8754d3bba44fe7f798aac7d0ffe2ca68fc75b92ff1a1ed29986ba7.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "0ccedd21865ba084d1ff6d8fd5398596d1f9405de58ee891ae343229eef3a433", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "a48b38fbd2981447a3f0e5de113f6ca5ff63aed4c4a58a72ab2f282fb616c495", "generation_seconds_batch": 25.769147370010614, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.043185263872146606, "target_z": -0.6171874759915607, "ce": 6.2767791748046875, "genuineness": 4.099404335021973, "identity": 0.3974355459213257, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/0ccedd21865ba084d1ff6d8fd5398596d1f9405de58ee891ae343229eef3a433.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "0cdd592ea45268caae192840a4b19e1abe6d70697aaaf1bc1c1a070d0f018652", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "683802879cec7f95d27444b22fdb07490e3d9a9491eced65d590baff7d534f0e", "generation_seconds_batch": 19.642417176044546, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.029435068368911743, "target_z": -0.47851564078332587, "ce": 5.808161735534668, "genuineness": 4.017816066741943, "identity": 0.7280091643333435, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/0cdd592ea45268caae192840a4b19e1abe6d70697aaaf1bc1c1a070d0f018652.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "0d0c7fab179b20cb8df740688fcc4a719d95f701c17561bd45187c5445dfc69f", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "48729192a292305af3fdf2b722e4d6850970e521814928557b362d99007ca58f", "generation_seconds_batch": 29.596475759055465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.043185263872146606, "target_z": -0.6171874759915607, "ce": 6.0146708488464355, "genuineness": 4.494794845581055, "identity": 0.5401554107666016, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/0d0c7fab179b20cb8df740688fcc4a719d95f701c17561bd45187c5445dfc69f.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "0d4761c3f0bc62015654e18e8ca923f7323111fc7531dfdb0ca2db0685832719", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "b98261ab6493b7bda71d2210d1053b20fc7b374e6b581a0eed9ecf046c8c835f", "generation_seconds_batch": 22.754951155977324, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.010001450777053833, "target_z": -0.5156250195624321, "ce": 5.982900142669678, "genuineness": 4.143336772918701, "identity": 0.6563888192176819, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/0d4761c3f0bc62015654e18e8ca923f7323111fc7531dfdb0ca2db0685832719.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": "0e0beb3fce1b2be25dd015c82a7b1771fd3326fee106df392b7259974593a743", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "791bfb81fbc2e085721908b53f3d908a0801aa54274c7d750d6ad199d2c83e1b", "generation_seconds_batch": 23.09301681793295, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04477740824222565, "target_z": -0.44921874333098905, "ce": 6.316492557525635, "genuineness": 4.331617832183838, "identity": 0.5416103601455688, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/0e0beb3fce1b2be25dd015c82a7b1771fd3326fee106df392b7259974593a743.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": "0ec5c899147d7e8f44695c88e5787e99e566c49dc59ff56826e297a500632d3c", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "08a3375a896376706396158355533d8beaed0018dcae9a76a8b5a9464bb86772", "generation_seconds_batch": 17.211229598033242, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.016138404607772827, "target_z": -0.5039062264361613, "ce": 6.173524379730225, "genuineness": 4.319066047668457, "identity": 0.8175176978111267, "hypothesis": "La route tourne à gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/0ec5c899147d7e8f44695c88e5787e99e566c49dc59ff56826e297a500632d3c.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": "0ecf57fb96dccecc67303172b1324b4f0a6fc69582eace715f98bb068c32c6c7", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e604056550a9b3edabf74247f8d1b69213cc10c73c58f688b0d583d703489a22", "generation_seconds_batch": 17.096285445033573, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03659482300281525, "target_z": -0.46484376289342116, "ce": 6.157639026641846, "genuineness": 4.545003414154053, "identity": 0.4914604127407074, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/0ecf57fb96dccecc67303172b1324b4f0a6fc69582eace715f98bb068c32c6c7.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": "0eeed76a931a0d81810d6a5aa86caf115177bb51bc7e7e3f44011821cd0d2c64", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "931dbc2d0d0293202e7bc23b1654b57798a50040bf54fa17729c7d050508ff63", "generation_seconds_batch": 17.19549462106079, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09694132208824158, "target_z": -0.34960937166549455, "ce": 6.2767791748046875, "genuineness": 3.9173994064331055, "identity": 0.41252797842025757, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, lies sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/0eeed76a931a0d81810d6a5aa86caf115177bb51bc7e7e3f44011821cd0d2c64.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "0fe298054ad907de6dc0b6de169cf14f3fa8361e4aa3491b305aaf54f8eadc7a", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "9f16678d0cad66dc9f8bf971f43acc63435f12974e73606ede71eaba629aaf7d", "generation_seconds_batch": 25.678026392008178, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8691718578338623, "target_z": 1.12500004268167, "ce": 6.221180438995361, "genuineness": 4.080576419830322, "identity": -0.051706697791814804, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/0fe298054ad907de6dc0b6de169cf14f3fa8361e4aa3491b305aaf54f8eadc7a.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "0ff5a93a72b4e0fe140ef7b26b768cca14f88ca337cdf3ccfee22bdca1c4c3b0", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "84b3e91fc5cf9a93df23d5f4dcf014d29df5d35b520576720a999bb35afe5794", "generation_seconds_batch": 18.677141251042485, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00636368989944458, "target_z": -0.5468750017784029, "ce": 6.260893821716309, "genuineness": 4.394378662109375, "identity": 0.725202739238739, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/0ff5a93a72b4e0fe140ef7b26b768cca14f88ca337cdf3ccfee22bdca1c4c3b0.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "1005971042395673d221f024d9dd4635bef392c7977879d0683e7feee4657e25", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "9695393ec37f5879667365c4df29c77775ff08061f10bf9c94e8cad8f3876d21", "generation_seconds_batch": 19.56057229801081, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03148071467876434, "target_z": -0.47460938589271784, "ce": 5.903473377227783, "genuineness": 4.13078498840332, "identity": 0.7010535597801208, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1005971042395673d221f024d9dd4635bef392c7977879d0683e7feee4657e25.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": "104e68914238bbf151b2770a2c12da05d2fd5e1ed8640bc47d69e68bb8110ff2", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a4656c82902296d1ead959fd458ab7b1bbc855cb25bc95fd5562f628c07ccc8c", "generation_seconds_batch": 27.323564617079683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01045498251914978, "target_z": -0.5546875115596189, "ce": 6.157639026641846, "genuineness": 4.068024158477783, "identity": 0.4371070861816406, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade, dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/104e68914238bbf151b2770a2c12da05d2fd5e1ed8640bc47d69e68bb8110ff2.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": "1058c29941cb79aa243c5502aa668c4aab874f3fa2997a7362efd16718cb3eaa", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Your message arrived after I had left.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b694dc82ae2843a213c0210bc0ac6c7719477f7f86812caef005751142c94aac", "generation_seconds_batch": 19.421222764067352, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0059101879596710205, "target_z": -0.5234374724347548, "ce": 6.229123115539551, "genuineness": 3.8169827461242676, "identity": 0.39299479126930237, "hypothesis": "Your message arrived after I had left. I read it twice, before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1058c29941cb79aa243c5502aa668c4aab874f3fa2997a7362efd16718cb3eaa.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "10a17c4a39f5ba20b37243c1c481780b55cac0fb96bf1cc3c029372443088b51", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "69b3226b5288f7efc85ffc166d46dc9626ef1a57374bd00b1ea7214a66c7b8cf", "generation_seconds_batch": 26.973854048992507, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.029435068368911743, "target_z": -0.47851564078332587, "ce": 5.613565921783447, "genuineness": 4.13078498840332, "identity": 0.5144517421722412, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/10a17c4a39f5ba20b37243c1c481780b55cac0fb96bf1cc3c029372443088b51.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": "10c9aa89634dfcdfb944876d71e4a4ba72191b145a7c1dd17bdf6e20df94da22", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1528a943765fbe282fb28d65fccd82571a7a26fc2618d6c3ea24b2e988ac15d4", "generation_seconds_batch": 17.218645321088843, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.025343775749206543, "target_z": -0.48632815056454193, "ce": 6.268836498260498, "genuineness": 4.356721878051758, "identity": 0.5828906893730164, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/10c9aa89634dfcdfb944876d71e4a4ba72191b145a7c1dd17bdf6e20df94da22.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "10d5241037166d1adc0161ba2d6e784a389ae80ec325c05e3aafda176e1b7ab2", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "399b6c8a1a520c107bde39b6a1525a05ca70e7d9bbe5452fb28c720c2341d333", "generation_seconds_batch": 26.76419600599911, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8446241021156311, "target_z": 1.0781249839943738, "ce": 6.252951145172119, "genuineness": 4.042920112609863, "identity": 0.06925724446773529, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/10d5241037166d1adc0161ba2d6e784a389ae80ec325c05e3aafda176e1b7ab2.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "11272fac16a041999d69fe5d6cb5919dfaea2ed96f181a04792f45306199e426", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "2d7ee547b29b3710e0bc2877051575dd21eb36ce60e045567bba44e3c08c8c3d", "generation_seconds_batch": 28.54341738205403, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01045498251914978, "target_z": -0.5546875115596189, "ce": 6.1894097328186035, "genuineness": 4.086852550506592, "identity": 0.7690898180007935, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/11272fac16a041999d69fe5d6cb5919dfaea2ed96f181a04792f45306199e426.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "116bced3fed80600c56f0c0194abcef7dea745126dc018c31cd7e1f7619c08cb", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "c13a74e0e58eb9cb27eb2d12a4655f703f15fb507944ce1786df44c395565295", "generation_seconds_batch": 25.118259939947166, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.008409321308135986, "target_z": -0.5507812282145643, "ce": 6.078212261199951, "genuineness": 4.425758361816406, "identity": 0.6194757223129272, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/116bced3fed80600c56f0c0194abcef7dea745126dc018c31cd7e1f7619c08cb.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "11c8317cb0b3cca0fccc2effaaedffa588c49cdec88eb652abfd784fd7470c88", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "636381d845ea9ec0903152bb92dcdf15d7119fd013462a76816abb060b71c68e", "generation_seconds_batch": 29.797443560906686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7402962446212769, "target_z": 0.8789061837544914, "ce": 6.205295085906982, "genuineness": 4.030367851257324, "identity": 0.5127685070037842, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/11c8317cb0b3cca0fccc2effaaedffa588c49cdec88eb652abfd784fd7470c88.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "1249405967955c7a692979ce3918a95c5f9ff2910ce2100ae984b9ea0cd94583", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "dedf119b586f22763854a914f0bdacf6a08841f2569def4c673c8336324c420c", "generation_seconds_batch": 26.32377220503986, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02432096004486084, "target_z": -0.4882812637826226, "ce": 6.0146708488464355, "genuineness": 4.080576419830322, "identity": 0.6686880588531494, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1249405967955c7a692979ce3918a95c5f9ff2910ce2100ae984b9ea0cd94583.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "1266af85d6f7650b1c8e20329540ce88be0381fbfcc773d203d3edf7e716c4e7", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "f9249b3bde2f937b800d0b8269ee0565df5ae7ef8f5264e7244a76c82b25562a", "generation_seconds_batch": 26.153902292018756, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00227239727973938, "target_z": -0.5390624919971869, "ce": 6.070269584655762, "genuineness": 4.331617832183838, "identity": 0.42522668838500977, "hypothesis": "Encontrado la cajita, estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.058823529411764705, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1266af85d6f7650b1c8e20329540ce88be0381fbfcc773d203d3edf7e716c4e7.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "1283e5f91ab03e321532e716751a1ef7c9a1eebdf9ac7793c1bc667c4822b3f9", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "196dbbffb9f34b091c5228ca2020f0e8be3cc89856b233098c7c3dbecf5acbe0", "generation_seconds_batch": 28.15937693894375, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00227239727973938, "target_z": -0.5390624919971869, "ce": 6.157639026641846, "genuineness": 4.262581825256348, "identity": 0.5733500123023987, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1283e5f91ab03e321532e716751a1ef7c9a1eebdf9ac7793c1bc667c4822b3f9.mp3" }, { "id": "heldout-es-0", "language": "es", "sentences": [ "Tu mensaje llegó después de que me fuera.", "Lo leí dos veces... y guardé el teléfono.", "Dime cuándo podemos hablar." ], "text": "Tu mensaje llegó después de que me fuera. Lo leí dos veces... y guardé el teléfono. Dime cuándo podemos hablar.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-0", "key": "12a7b850c68b1e935d69d2b179f3c0df54c4775061935760902fd9f6cbfc7541", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Intense Malevolence Malice. 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": "85c35da96b21ead877f6902f595b1b3fbb5c096d1d19d7e62097b36f4909eae1", "generation_seconds_batch": 19.012101776083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.041708946228027344, "target_z": -0.45507811143967775, "ce": 6.197352409362793, "genuineness": 4.287685871124268, "identity": 0.7613556385040283, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/12a7b850c68b1e935d69d2b179f3c0df54c4775061935760902fd9f6cbfc7541.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "12eef8d14195d62c42c0c072fd23c5566dc3154b68e672f55196252fd68f996e", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "53735254520420894dc59a0de9e5f7fe4a6e75998597f497ea7b1c88b90c7d49", "generation_seconds_batch": 23.972444336046465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.012047111988067627, "target_z": -0.5117187362173774, "ce": 6.24500846862793, "genuineness": 4.494794845581055, "identity": 0.7608656883239746, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/12eef8d14195d62c42c0c072fd23c5566dc3154b68e672f55196252fd68f996e.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "131a7b5a819ebd7509575d1d5c36d10fe6e62539739b9fc690e54b4a011aedd6", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "f2fb4d3997946bc7c71364ead27dd419f7ec6d10b94eff2f6a90f4626fc8c17f", "generation_seconds_batch": 26.350685394951142, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.012047111988067627, "target_z": -0.5117187362173774, "ce": 6.038498878479004, "genuineness": 4.20609712600708, "identity": 0.6001153588294983, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/131a7b5a819ebd7509575d1d5c36d10fe6e62539739b9fc690e54b4a011aedd6.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "13507e5310d8d5cc7ca06c1205f06bf83b2e74eb9217fb7e8cb3e0154b8860f2", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a04709c80b7e322b486b99eeef8b02130dc8904c4ae0301f710a5b243b33df93", "generation_seconds_batch": 29.06242644006852, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08159899711608887, "target_z": -0.3789062406633847, "ce": 6.38003396987915, "genuineness": 3.948779582977295, "identity": 0.5802655220031738, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/13507e5310d8d5cc7ca06c1205f06bf83b2e74eb9217fb7e8cb3e0154b8860f2.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "1361c5323822e9257a12e9ba33aa36adabe285f697445cc8e3017e4ef4fa465c", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2312577e4bd4f4e9bc2ff9dd6f60c70fa519496685203c8d0d159a20588be7c7", "generation_seconds_batch": 24.177027603029273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6710641384124756, "target_z": 2.656249911079854, "ce": 6.252951145172119, "genuineness": 3.9927117824554443, "identity": 0.3370566666126251, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1361c5323822e9257a12e9ba33aa36adabe285f697445cc8e3017e4ef4fa465c.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": "13d440155aff5c5934a5919dbf9365012ac80e88bbf7d1d59c88e7ff10299d25", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0415e4c8adfb8ad67eb9f2a3bbb02b9b940156b91b14259607d729ba4af42c0e", "generation_seconds_batch": 20.663879070896655, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0059101879596710205, "target_z": -0.5234374724347548, "ce": 6.205295085906982, "genuineness": 4.344170093536377, "identity": 0.6266550421714783, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna, naast het raam, neem ze mee als je weggaat.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/13d440155aff5c5934a5919dbf9365012ac80e88bbf7d1d59c88e7ff10299d25.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "14a630551af052a9f887827af501e20410b6ac09dbb8c4c43ae80113a28b6e47", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "cefae5fee420f160e15e09aeeb4c16e170036f72a402d413e5653ae2c3ddf948", "generation_seconds_batch": 28.102338333963417, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.04727655649185181, "target_z": -0.6249999857727767, "ce": 6.41180419921875, "genuineness": 4.2751336097717285, "identity": 0.3087732195854187, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/14a630551af052a9f887827af501e20410b6ac09dbb8c4c43ae80113a28b6e47.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "14bc2b5965b10828059ef1468b0915ef745d4639c69689edfb78eaa010a41a73", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3a60453197f0030c3d90d19c3bface2bb1e68f549739ac6e6569ef95ef40b5e7", "generation_seconds_batch": 27.50013582292013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.16240191459655762, "target_z": -0.22460938589271787, "ce": 6.173524379730225, "genuineness": 4.105680465698242, "identity": 0.5059338212013245, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/14bc2b5965b10828059ef1468b0915ef745d4639c69689edfb78eaa010a41a73.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": "1514e1c416915b31a45b0a47f5a063a38be5aed6983a3ce100859062e76204f1", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0d662a4f189ffc2125e71a961ca213006692bc74ee01a707dd9c7d9ecce57d99", "generation_seconds_batch": 19.938679017941467, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.026820123195648193, "target_z": -0.5859374937755898, "ce": 5.998785495758057, "genuineness": 4.143336772918701, "identity": 0.6429053544998169, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1514e1c416915b31a45b0a47f5a063a38be5aed6983a3ce100859062e76204f1.mp3" }, { "id": "heldout-de-2", "language": "de", "sentences": [ "Die Straße führt hinter der Brücke nach links.", "Geh weiter... bis zu dem kleinen Holztor.", "Ich warte auf der anderen Seite." ], "text": "Die Straße führt hinter der Brücke nach links. Geh weiter... bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-2", "key": "15585f3962d31fd159bb39a755593ab24aa0df87e4725d7772f40811258760f5", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Intense Malevolence Malice. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "267e0817fb4d5a61718fec3598006ad0e6feec8294e8ef5af890fc118fc5c07c", "generation_seconds_batch": 20.43981575698126, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00636368989944458, "target_z": -0.5468750017784029, "ce": 6.427689552307129, "genuineness": 4.105680465698242, "identity": 0.5379078984260559, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/15585f3962d31fd159bb39a755593ab24aa0df87e4725d7772f40811258760f5.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "1581ddc7f26a8fa7486c6f3cf73abd3b06685218f30bb398294ab173cc9ba342", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "a580832cabb269cdc4658f7ca244c78328ed71e9bd2d4ca40ad454785ddf9b46", "generation_seconds_batch": 24.778106614015996, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.016138404607772827, "target_z": -0.5039062264361613, "ce": 5.967014789581299, "genuineness": 4.1998209953308105, "identity": 0.31644123792648315, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1581ddc7f26a8fa7486c6f3cf73abd3b06685218f30bb398294ab173cc9ba342.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": "15e0874e0a2dbd892956a291ab782ff1e9916fe6e3994b4bd33701d097fc3261", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Someone moved the little blue chair.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "71edcc40d09838947da7faeec0b73e3206c52bf6b8adcd8394d3ba4f02362c64", "generation_seconds_batch": 27.68157036800403, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.008409321308135986, "target_z": -0.5507812282145643, "ce": 6.332377910614014, "genuineness": 4.457139015197754, "identity": 0.12593673169612885, "hypothesis": "Someone moved the little blue chair. It used to stand here beside the tall bookshelf. We can put it back after lunch.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/15e0874e0a2dbd892956a291ab782ff1e9916fe6e3994b4bd33701d097fc3261.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "163701edf9da4003a9188c6906aa80f21432b33e18a80227304f310ed43345c3", "bucket": "emo_Malevolence_Malice", "phase": "wording", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "28bbfc6e4acd46ab7d4a4686d167a9c686dd9d0dc53da06f44512b445160d7c4", "generation_seconds_batch": 21.299496843013912, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04580022394657135, "target_z": -0.4472656301129084, "ce": 6.284721851348877, "genuineness": 4.394378662109375, "identity": 0.7141426801681519, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/163701edf9da4003a9188c6906aa80f21432b33e18a80227304f310ed43345c3.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "1649319436efc58a4050cc7f993a77d63b5f7a99ab7d9c164a8a5c0f7aff5f2b", "bucket": "emo_Malevolence_Malice", "phase": "wording", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "81cb4bf38e7a73d694c10ad3d2fd4b815c241a213e3c03af768ff5e52cd3c35b", "generation_seconds_batch": 20.27062622201629, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.025343775749206543, "target_z": -0.48632815056454193, "ce": 6.0146708488464355, "genuineness": 3.8546390533447266, "identity": 0.41921502351760864, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1649319436efc58a4050cc7f993a77d63b5f7a99ab7d9c164a8a5c0f7aff5f2b.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": "16e962d41f08a6360bc97f2ffbd543c8c4f019e66afd699d20cde691c1bbb982", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "3a5fd8893b92e8e6d789a7ba4a7409aeb25b5b7196c7a3a01b17e46ac13959c2", "generation_seconds_batch": 22.108712989953347, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.0452309250831604, "target_z": -0.6210937593366153, "ce": 6.1020402908325195, "genuineness": 4.331617832183838, "identity": 0.6379992961883545, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/16e962d41f08a6360bc97f2ffbd543c8c4f019e66afd699d20cde691c1bbb982.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "1817d22624470266c4641c7d478215ff6daa8cc2615abcc4e09076bc3846af44", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "aecf7fa375105ca8d5f87324c7f95d29d909b61f82a6c7eac3276acccc987d7e", "generation_seconds_batch": 29.113067654077895, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06727948784828186, "target_z": -0.40624999644319415, "ce": 6.09409761428833, "genuineness": 3.9425034523010254, "identity": 0.3587316870689392, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1817d22624470266c4641c7d478215ff6daa8cc2615abcc4e09076bc3846af44.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "182217a282de15b785dafa7559c3bbe1a448a30b371e262b967bfeabeb76fadb", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "d6d8c3f055f1a1de62bd3633ad1173097f67df8b1a883ca7fe70fb99b6bc3eea", "generation_seconds_batch": 26.026272892020643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.055005624890327454, "target_z": -0.42968749733239564, "ce": 6.086154937744141, "genuineness": 3.8295347690582275, "identity": 0.1608055830001831, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/182217a282de15b785dafa7559c3bbe1a448a30b371e262b967bfeabeb76fadb.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "1880d0e6c4f650895991abb02db3673911e097df0960527a65ec5a7f00e99503", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "a232ec032f34c7902138c94d4860028af327915340e5ab928b180f15c52d0ea0", "generation_seconds_batch": 18.716593179968186, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00227239727973938, "target_z": -0.5390624919971869, "ce": 6.157639026641846, "genuineness": 4.218649387359619, "identity": 0.6741175055503845, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1880d0e6c4f650895991abb02db3673911e097df0960527a65ec5a7f00e99503.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": "18b30ce85d9f34be488154808f845635ef81ef8dfb9e11bad8e2ee6b01780b2b", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7f87e181eb665bd5d5492cdcf81fe21c64f9e03f6abff959a78df84e5912d601", "generation_seconds_batch": 17.199707175022922, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.022728830575942993, "target_z": -0.5781249839943737, "ce": 6.157639026641846, "genuineness": 4.281409740447998, "identity": 0.7372477054595947, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue. Elle était ici, près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/18b30ce85d9f34be488154808f845635ef81ef8dfb9e11bad8e2ee6b01780b2b.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "19234b41ef9ca245c895b0c0ae8f149816980f06f525ae5ac28399d98b63064e", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1a50d03a530a0b9a0be23ea9a1f5a89cb6722457c88a51ef4b4fa2f38a995aee", "generation_seconds_batch": 28.461921441950835, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08262181282043457, "target_z": -0.376953127445304, "ce": 6.300607204437256, "genuineness": 4.356721878051758, "identity": 0.2334502637386322, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/19234b41ef9ca245c895b0c0ae8f149816980f06f525ae5ac28399d98b63064e.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "192aee46f2a76ccfc7be53ef38334607899ed4a3235bd5415379b3bd5019c769", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "6b07ec313302caa32486fea52d492a5b8167c1c3fcc2a6814efa0848964f2c1a", "generation_seconds_batch": 26.20452650100924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.15677639842033386, "target_z": -0.23535156550105493, "ce": 5.788304805755615, "genuineness": 3.973883628845215, "identity": 0.6424610614776611, "hypothesis": "De deur staat nog open? Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/192aee46f2a76ccfc7be53ef38334607899ed4a3235bd5415379b3bd5019c769.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "193858ab11ffedf09b080a309f58f61807e7f1725a216e8ec92bf42a04902d17", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": 127, "seconds_generated": 10.16, "mp3_sha256": "0b4be9211090ccfc1fc1a84ef7d7c276d378069d4bf0af45d3fba3374a1da761", "generation_seconds_batch": 21.528846655972302, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.022728830575942993, "target_z": -0.5781249839943737, "ce": 6.260893821716309, "genuineness": 4.21237325668335, "identity": 0.612450361251831, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/193858ab11ffedf09b080a309f58f61807e7f1725a216e8ec92bf42a04902d17.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "19479277edcf4945ac1be5013eb9481e29a790f38199576912d2f81dbad1898e", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "f7fe35fb62def079b70d9aa2a2572a074b98b83d4dd5bc7b4a9812c1b2394818", "generation_seconds_batch": 25.002688780077733, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.19717785716056824, "target_z": -0.1582031381157215, "ce": 5.895530700683594, "genuineness": 4.18099308013916, "identity": 0.013991749845445156, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/19479277edcf4945ac1be5013eb9481e29a790f38199576912d2f81dbad1898e.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "197630ad01ac25546b92c9eb8611d74d50348bf7c8c9008449e1cd1ee159de08", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "d099bf4119d2e9c72f1f507164d28bdc2768016cfdd279f6b0a061c0a2d3b763", "generation_seconds_batch": 25.78369962598663, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.018184036016464233, "target_z": -0.5, "ce": 6.197352409362793, "genuineness": 4.319066047668457, "identity": 0.36409682035446167, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/197630ad01ac25546b92c9eb8611d74d50348bf7c8c9008449e1cd1ee159de08.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "19c91601db9fcd5580c9daf6d69b7aa49cc40156a37af51be88d686902f2a6fe", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "54f3bc31295624604351a49aced514b2afc50b3804f637f43cd6d7ab26b8f195", "generation_seconds_batch": 28.979571270989254, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.028865784406661987, "target_z": -0.5898437771206445, "ce": 6.268836498260498, "genuineness": 4.068024158477783, "identity": 0.2099420428276062, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/19c91601db9fcd5580c9daf6d69b7aa49cc40156a37af51be88d686902f2a6fe.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "1a3465ee576e1285abe94125bface5572514f11e3054ad45273649d2a50f908d", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "ea977f6542f2c83679f5a491c962c822a1035f6e2f2b8f8daa6bd832662f7295", "generation_seconds_batch": 25.800091105978936, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03659482300281525, "target_z": -0.46484376289342116, "ce": 6.125868320465088, "genuineness": 4.482243061065674, "identity": 0.6584697961807251, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1a3465ee576e1285abe94125bface5572514f11e3054ad45273649d2a50f908d.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": "1a468e6f2c4f71d8f6cbbfb56575684e086edbcf7bcdb9a0b084b09d24afeefe", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Escribí el número en el sobre.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "24273acda59e297e6b8f317e839e3fa237984ba41c97ea41f1eeda0b9229f574", "generation_seconds_batch": 17.214052723022178, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.059096917510032654, "target_z": -0.4218749875511796, "ce": 6.09409761428833, "genuineness": 4.268857479095459, "identity": 0.6483097672462463, "hypothesis": "Escribí el número en el sobre, léelo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1a468e6f2c4f71d8f6cbbfb56575684e086edbcf7bcdb9a0b084b09d24afeefe.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "1a8cea27106db35dc40644e51e34f29b32c985374c88d07ad0a8b860721ec049", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "06b439b983550c8bc2e3713a6ad5151258641d5c9784de6d79941e13fa2db92f", "generation_seconds_batch": 27.355432596057653, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.03704833984375, "target_z": -0.6054687397741833, "ce": 6.086154937744141, "genuineness": 4.2751336097717285, "identity": 0.6146178245544434, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1a8cea27106db35dc40644e51e34f29b32c985374c88d07ad0a8b860721ec049.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": "1aa190acfd7b8d3d9cef5f2ef2fcb9dbfe947cefabf7b2c66acbf671bc498b2f", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9528bf6a3c7b8dad08468bdd3239958eb4bc77bef8c6eea348a02bbfcfe35344", "generation_seconds_batch": 22.320822505047545, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.020683199167251587, "target_z": -0.5742187575582124, "ce": 6.09409761428833, "genuineness": 4.419482707977295, "identity": 0.7167096138000488, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue. Elle était ici, près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1aa190acfd7b8d3d9cef5f2ef2fcb9dbfe947cefabf7b2c66acbf671bc498b2f.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": "1ac3acffca5e2ae5fc3350bfff4186c699718233eb038c06fa45fd9fa3f5a68c", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "221e33ba9711cfb813cd937ccc02306266c3a3e6d903f8ef91c02d1d463d7c81", "generation_seconds_batch": 17.218645321088843, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.037617653608322144, "target_z": -0.4628906212208938, "ce": 6.157639026641846, "genuineness": 4.055472373962402, "identity": 0.7186641097068787, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1ac3acffca5e2ae5fc3350bfff4186c699718233eb038c06fa45fd9fa3f5a68c.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "1b1dc5c6d1fdc388b7685230b573e0ea38290e4249b64fa663e0d20b1215b8d5", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0acc513b1507eab5f158c6ebec3b1a4fc3ebbf61357039bef0420140dc66a535", "generation_seconds_batch": 27.70254610606935, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08773592114448547, "target_z": -0.3671875044460073, "ce": 5.982900142669678, "genuineness": 4.15588903427124, "identity": 0.47801482677459717, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1b1dc5c6d1fdc388b7685230b573e0ea38290e4249b64fa663e0d20b1215b8d5.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "1b34bdeac9c1933c91c460dce5258320744c7d3691b721bc94a6f257ad4df6ae", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "a755e0de88141e493ca6f0f9482cb710a75a2771472e224b202f21a730504cb9", "generation_seconds_batch": 26.237071320065297, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1245575100183487, "target_z": -0.2968750017784029, "ce": 6.2767791748046875, "genuineness": 3.9801597595214844, "identity": 0.27866417169570923, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1b34bdeac9c1933c91c460dce5258320744c7d3691b721bc94a6f257ad4df6ae.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "1b6f010f0c336a88f3adc84ca52d231aaf72ff43cc04f4cc62d3ddbb36c59740", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "930f48725fe4316ff7cb4b3ca559e191c7a6dff70fd0a0c6e16851024c2e816c", "generation_seconds_batch": 28.741010984987952, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00227239727973938, "target_z": -0.5390624919971869, "ce": 6.332377910614014, "genuineness": 4.507347106933594, "identity": 0.36358505487442017, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1b6f010f0c336a88f3adc84ca52d231aaf72ff43cc04f4cc62d3ddbb36c59740.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "1b811398b22c57a0c1bd27810efe02261e4e11ede1c108806114279530f2596c", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "1d899223a2144cf1c3ac22dc67b5a73290243f9608529c7fd444d7097d02dc78", "generation_seconds_batch": 26.92806115502026, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.06773298978805542, "target_z": -0.6640624777699635, "ce": 6.332377910614014, "genuineness": 4.432034492492676, "identity": 0.5079603791236877, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1b811398b22c57a0c1bd27810efe02261e4e11ede1c108806114279530f2596c.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "1c55d245012c23985c0b67c0091f68f0e1caf482baf958fd502296457c5defb3", "bucket": "emo_Malevolence_Malice", "phase": "wording", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "90a451b89e279c017ce8fe3a17047cf4f051220177e490c39526acc80cb103e6", "generation_seconds_batch": 19.104797292035073, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02125251293182373, "target_z": -0.49414060343686467, "ce": 6.109982967376709, "genuineness": 4.2249250411987305, "identity": 0.6141990423202515, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1c55d245012c23985c0b67c0091f68f0e1caf482baf958fd502296457c5defb3.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "1c73312c126f750c776e6673af6eb798671de29a2830c4372799ecb37f829fb8", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "25927d305e8c4e4d857ee34acdbcde955f59b7b4e6d0081972e76ce465edc2c6", "generation_seconds_batch": 19.622488979948685, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.06364172697067261, "target_z": -0.6562500248976408, "ce": 6.427689552307129, "genuineness": 4.143336772918701, "identity": 0.2577524781227112, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1c73312c126f750c776e6673af6eb798671de29a2830c4372799ecb37f829fb8.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "1d2bc07bdcec9ebb76cb1169b49f8f2479a1548230d5c1d6d0ef049fe175eabc", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "97e7e94ffa8dc81b8bca4509da845e17187b2e990eb5eba7ac3953e0889fa3e8", "generation_seconds_batch": 28.988100476097316, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03045789897441864, "target_z": -0.47656249911079857, "ce": 5.990842819213867, "genuineness": 4.15588903427124, "identity": 0.5136234164237976, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1d2bc07bdcec9ebb76cb1169b49f8f2479a1548230d5c1d6d0ef049fe175eabc.mp3" }, { "id": "heldout-es-4", "language": "es", "sentences": [ "Escribí el número en el sobre.", "Léelo bien... y comprueba las dos últimas cifras.", "Llámame cuando tengas un momento." ], "text": "Escribí el número en el sobre. Léelo bien... y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-4", "key": "1d559e6da25f8bbacc3b53d5880369445633ce7ab0125ca7b07093108b8c8de3", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Escribí el número en el sobre.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Escribí el número en el sobre.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Intense Malevolence Malice. 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": "8d7096087db5851832581279f951f9bff5c79e8cae86c4c6a28f1022a24c117c", "generation_seconds_batch": 20.612828240031376, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04580022394657135, "target_z": -0.4472656301129084, "ce": 6.300607204437256, "genuineness": 4.3881025314331055, "identity": 0.4952417016029358, "hypothesis": "Escribí el número en el sobre, léelo bien, y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1d559e6da25f8bbacc3b53d5880369445633ce7ab0125ca7b07093108b8c8de3.mp3" }, { "id": "heldout-de-1", "language": "de", "sentences": [ "Die Schlüssel liegen auf dem Küchentisch.", "Ich suchte in der Schublade... dann neben dem Fenster.", "Nimm sie mit, wenn du gehst." ], "text": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade... dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-1", "key": "1d829402ab2176d2024ae3e918286a39ec094d9a5299caa0e048ba37912e1579", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Intense Malevolence Malice. 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": "6827bb07f57b32e91c3c86690495ff51972c6498f9a714b21f4f14eb11552fe0", "generation_seconds_batch": 18.91091773903463, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.014092743396759033, "target_z": -0.5078125097812161, "ce": 6.38003396987915, "genuineness": 4.344170093536377, "identity": 0.7428750395774841, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade, dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1d829402ab2176d2024ae3e918286a39ec094d9a5299caa0e048ba37912e1579.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": "1de53f0cb5a35f8c05c40196accbae8d7410c0a99fb74e4d09fddba02dad15fb", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3d845f3a659422c248bdabf7afad6985391a0bfedfc95c17064cac446d84a1f1", "generation_seconds_batch": 22.176203629001975, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.018184036016464233, "target_z": -0.5, "ce": 6.23706579208374, "genuineness": 4.18099308013916, "identity": 0.4411880075931549, "hypothesis": "La route tourne à gauche après le pont, continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1de53f0cb5a35f8c05c40196accbae8d7410c0a99fb74e4d09fddba02dad15fb.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "1ea808a265d1867542acdfea8eb1c2b217d6512b0225726af4c83a83b3b30fc4", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "945b97cbc127b04085302e0dc08d033e3691231d0a48fde0c7076b18a8b86539", "generation_seconds_batch": 27.325723235029727, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00022676587104797363, "target_z": -0.5351562655610255, "ce": 5.903473377227783, "genuineness": 3.7040140628814697, "identity": 0.5292536020278931, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1ea808a265d1867542acdfea8eb1c2b217d6512b0225726af4c83a83b3b30fc4.mp3" }, { "id": "heldout-es-4", "language": "es", "sentences": [ "Escribí el número en el sobre.", "Léelo bien... y comprueba las dos últimas cifras.", "Llámame cuando tengas un momento." ], "text": "Escribí el número en el sobre. Léelo bien... y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-4", "key": "1eaf0e0e2c966186cf6ada2de41819a315f8b1b1eeeeef5a84a28dd582998db8", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Escribí el número en el sobre.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Escribí el número en el sobre.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Intense Malevolence Malice. 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": "570c3495bf3ed1062ce65e51947235e9ec9345b1665b5d2701134fd21986438b", "generation_seconds_batch": 20.612828240031376, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0038645267486572266, "target_z": -0.5273437557798095, "ce": 6.340320587158203, "genuineness": 4.3127899169921875, "identity": 0.6552132368087769, "hypothesis": "Escribí el número en el sobre. Léelo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1eaf0e0e2c966186cf6ada2de41819a315f8b1b1eeeeef5a84a28dd582998db8.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "1ec68be2594b25683b121d82176d64873638d4c51567c3a16ea94fb5a77df4b1", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "c0c6aafc0c3b2f2b15feefd68edddc173391e990f3d95563f42535f0088650fa", "generation_seconds_batch": 21.186607319978066, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.33679306507110596, "target_z": 0.10839846651019756, "ce": 6.3641486167907715, "genuineness": 4.143336772918701, "identity": 0.5206155776977539, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1ec68be2594b25683b121d82176d64873638d4c51567c3a16ea94fb5a77df4b1.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": "1f5529184e625e22bb5644e1da950482ff00d197bd8d07d86cd86b0940fd83b8", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b06619805de961385db8ad3c02276d129d04b2e8426f42630eb43e2d75168ebb", "generation_seconds_batch": 18.956564913038164, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10103261470794678, "target_z": -0.3417968618842785, "ce": 6.221180438995361, "genuineness": 4.344170093536377, "identity": 0.44890540838241577, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1f5529184e625e22bb5644e1da950482ff00d197bd8d07d86cd86b0940fd83b8.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "1fad3418205a36a4f79731c1ba1d0144d7ece3b38f35d8b9d94ef0319099574d", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "b62222ba21ebdc3d84208ec986b714ed18daf00ea82897643525439efd1a1599", "generation_seconds_batch": 27.835458765970543, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5786904096603394, "target_z": 0.5703124742131577, "ce": 6.332377910614014, "genuineness": 4.2751336097717285, "identity": 0.44613322615623474, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1fad3418205a36a4f79731c1ba1d0144d7ece3b38f35d8b9d94ef0319099574d.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": "1fd17c5ae9c50c44a7448ed445b438a9eb77a0402af016f6564079dff5433f8c", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "81dcea393726fea390ec8858f3a384b087c4ba5b50d5137262a4b8ee3367f97b", "generation_seconds_batch": 19.938679017941467, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.012500613927841187, "target_z": -0.5585937379957803, "ce": 6.046441555023193, "genuineness": 4.582659721374512, "identity": 0.46132129430770874, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai regardé dans le tiroir, puis près de la fenêtre. Prends les quand tu partiras.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1fd17c5ae9c50c44a7448ed445b438a9eb77a0402af016f6564079dff5433f8c.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "1fe7f4cbbeb57ca3aa361bedadb19532f69fc160a9166434860302db1a8b43de", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "bcc3b728c3b8ded29e37b3bab390cb4d442f7884aa931fbbd3f75de87e4e5c52", "generation_seconds_batch": 20.465618073008955, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03148071467876434, "target_z": -0.47460938589271784, "ce": 6.292664527893066, "genuineness": 4.3881025314331055, "identity": 0.5637973546981812, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/1fe7f4cbbeb57ca3aa361bedadb19532f69fc160a9166434860302db1a8b43de.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "202a998ea908d4f0359098ec8adcbafc0f9dc9dd2cb82d63fda3ce26dde6d37d", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f38ebd965aa88c9096be876111b8b087fe17cee88b138478073969a3df0f6f82", "generation_seconds_batch": 25.024099898990244, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.16137909889221191, "target_z": -0.22656249911079854, "ce": 6.316492557525635, "genuineness": 4.15588903427124, "identity": 0.4156421422958374, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/202a998ea908d4f0359098ec8adcbafc0f9dc9dd2cb82d63fda3ce26dde6d37d.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "206aff0dfac8d4c4d2e90045f7b708263d83d03642b6c45324f308121f990b71", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "de98a8a409b061e521c255598eb1120af27984d09d8af4ae2372799f39c9c5a3", "generation_seconds_batch": 19.285099836066365, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.12353469431400299, "target_z": -0.2988281149964836, "ce": 6.165581703186035, "genuineness": 4.413206577301025, "identity": 0.7596423625946045, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/206aff0dfac8d4c4d2e90045f7b708263d83d03642b6c45324f308121f990b71.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "2114953030a36808dc162e7946bf0580dfbda264f56da7bf6a165423eba4ebee", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "3788ac4fc618b20e8b31f10f0176fd163cff01d44a6ae8b7d805d8e21e364a7f", "generation_seconds_batch": 33.55481519794557, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.010001450777053833, "target_z": -0.5156250195624321, "ce": 6.260893821716309, "genuineness": 4.432034492492676, "identity": 0.681925356388092, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/2114953030a36808dc162e7946bf0580dfbda264f56da7bf6a165423eba4ebee.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": "2196f7263bdd95cb37ba4cd82c8ac62228eddcfadd7b99e794772afc604b2088", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3cf5817c039b532b1cff60b71cd33b0e243949ba90709fd3ce3b3dd5f192a045", "generation_seconds_batch": 17.240327014005743, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.33986151218414307, "target_z": 0.11425780616443962, "ce": 6.316492557525635, "genuineness": 4.268857479095459, "identity": 0.4654029905796051, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/2196f7263bdd95cb37ba4cd82c8ac62228eddcfadd7b99e794772afc604b2088.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": "2210444f0e2b12fa2b4fc7687e233e7ac5ddc94620e931ea8a822affbdb04dcf", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aecb27462d6daebe2c765d1d55211eb3af7db13a8f022739aed74bbf48d69846", "generation_seconds_batch": 22.4599160089856, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05193716287612915, "target_z": -0.43554686544108434, "ce": 6.181467056274414, "genuineness": 4.13078498840332, "identity": 0.4031279683113098, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/2210444f0e2b12fa2b4fc7687e233e7ac5ddc94620e931ea8a822affbdb04dcf.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": "222fd3ac9f4880fb848645971981f03e25820eaeab347a64c28ed6b114a592c1", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "226e0a46666a3f0fcff580070adea01a7738ac63ef157cdc2f579e3eb43c5ec5", "generation_seconds_batch": 17.272515240008943, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06625665724277496, "target_z": -0.4082031381157215, "ce": 6.054384231567383, "genuineness": 4.507347106933594, "identity": 0.6298583745956421, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/222fd3ac9f4880fb848645971981f03e25820eaeab347a64c28ed6b114a592c1.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "226ee215a7d8c927d7c4b2b68d19adde89d491eb12a4420bd5f082d9cf01671b", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "b6e5668ab5271ca7e3292866ca0b8a1e4aeca860f38ab0fea646c4eb19e31ca3", "generation_seconds_batch": 25.10108997800853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03045789897441864, "target_z": -0.47656249911079857, "ce": 6.213237762451172, "genuineness": 4.344170093536377, "identity": 0.6337790489196777, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/226ee215a7d8c927d7c4b2b68d19adde89d491eb12a4420bd5f082d9cf01671b.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": "22b985c705c395fad3dc1bab431a540906fb31ecc58d53d77d49c60d7e2bc257", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "216d9c9ea5cad84d490c774828fd5ff79d6007fe57ef2f6805cb09e95d572501", "generation_seconds_batch": 23.071617268957198, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0018188953399658203, "target_z": -0.5312499822159709, "ce": 6.252951145172119, "genuineness": 4.168440818786621, "identity": 0.5754953622817993, "hypothesis": "Ik schreef het nummer op de envelop, lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/22b985c705c395fad3dc1bab431a540906fb31ecc58d53d77d49c60d7e2bc257.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "22ce04846eb7f6e0b44ab1733bcba919b65dd4502b79c2a51e2a7004f6358393", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "324ac07b17276eb7e423c253fcc42a28486a9572891dd1042b9cfe49006de962", "generation_seconds_batch": 30.710311016999185, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08057616651058197, "target_z": -0.38085938233591204, "ce": 6.38797664642334, "genuineness": 4.093128204345703, "identity": 0.318467915058136, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/22ce04846eb7f6e0b44ab1733bcba919b65dd4502b79c2a51e2a7004f6358393.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": "22ec5ccc75467edbc880dc3db7b9d7fc39f66310ad877c838a1b3bcf5c5d3e7f", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8304a026ac11965b6d67ac44be059681033303326b5fc208ece142bf41715fba", "generation_seconds_batch": 17.175388882984407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00636368989944458, "target_z": -0.5468750017784029, "ce": 6.24500846862793, "genuineness": 3.986435890197754, "identity": 0.731407880783081, "hypothesis": "Las llaves están en la mesa de la cocina. Miré en el cajón y luego, junto a la ventana, llévatelas cuando te vayas.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/22ec5ccc75467edbc880dc3db7b9d7fc39f66310ad877c838a1b3bcf5c5d3e7f.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "238fe27df6d14622f0202d4b102fd46bb14ac1111d3e86c7bf09069fef5f31f6", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "07a022591186f8c3ce2216dd1cccc2f248b3581565b7d823750c396a95c480ce", "generation_seconds_batch": 25.1878926520003, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.018184036016464233, "target_z": -0.5, "ce": 6.133810997009277, "genuineness": 4.582659721374512, "identity": 0.5202969312667847, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/238fe27df6d14622f0202d4b102fd46bb14ac1111d3e86c7bf09069fef5f31f6.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "239f5a0dffe7f9e4005d905592bbfb2f14e48d8702411ee309e375ad74100000", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "518cd258f55f62a65326aff595e1a08eccc34f04e7539118affaeebc14a79cb3", "generation_seconds_batch": 20.142297467100434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.020683199167251587, "target_z": -0.5742187575582124, "ce": 6.268836498260498, "genuineness": 4.457139015197754, "identity": 0.6002058386802673, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/239f5a0dffe7f9e4005d905592bbfb2f14e48d8702411ee309e375ad74100000.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "23d22c0d6c270d501d9f00ee0ae157a40d668d971b8641c7e20f0889810705d5", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "33ebcff1214119db6321bd046a8767b552ee32079ead35b2482be9ae593ff769", "generation_seconds_batch": 26.110362123930827, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6298315525054932, "target_z": 0.6679688180239115, "ce": 6.316492557525635, "genuineness": 4.250029563903809, "identity": 0.4949249029159546, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/23d22c0d6c270d501d9f00ee0ae157a40d668d971b8641c7e20f0889810705d5.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "241ab872b3acb4bfa22cc9d83e70808b09eae22befea8d74e0810490ec546115", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "ad23796d6dd6de35c1d90a5a2a5bcaa26e2e45ee19e369265692dda9afde5a7a", "generation_seconds_batch": 20.851765237981454, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00227239727973938, "target_z": -0.5390624919971869, "ce": 5.919358730316162, "genuineness": 4.074300289154053, "identity": 0.2788885831832886, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/241ab872b3acb4bfa22cc9d83e70808b09eae22befea8d74e0810490ec546115.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "24f58ca9a544b54ff60c6698a56116593d44baaa6d54f4d53a76c234a56a0aad", "bucket": "emo_Malevolence_Malice", "phase": "wording", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "7fab8c43ade289719d8eb88fab9465f522e12f7b2227a0184df7e7ccddf826ff", "generation_seconds_batch": 20.429856966948137, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.028865784406661987, "target_z": -0.5898437771206445, "ce": 6.332377910614014, "genuineness": 4.532451152801514, "identity": 0.4229464530944824, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/24f58ca9a544b54ff60c6698a56116593d44baaa6d54f4d53a76c234a56a0aad.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "253a496bf54d6b33304172dd974c45b0757da097bdfdfe953f3703f1b1b45980", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "8222ad55186cfc8ee71f79c06061a5d424b239d3ef621c65389ff7e2b6e95f23", "generation_seconds_batch": 19.52777700603474, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05602844059467316, "target_z": -0.42773438411431497, "ce": 6.1020402908325195, "genuineness": 4.193545341491699, "identity": 0.25690796971321106, "hypothesis": "The door IS still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/253a496bf54d6b33304172dd974c45b0757da097bdfdfe953f3703f1b1b45980.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "2569a2d5fea06920365e858d6e58099babda97907633847aba9115cbc7cea3f0", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "ae9d605d310129321a41d6222c717ea73fa36f517bd426e8652fbfeff5afd5bf", "generation_seconds_batch": 26.92806115502026, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.0411396324634552, "target_z": -0.6132812495553993, "ce": 6.356205940246582, "genuineness": 4.331617832183838, "identity": 0.4330582618713379, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/2569a2d5fea06920365e858d6e58099babda97907633847aba9115cbc7cea3f0.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "25f7dd4083be03f05c73a9806a6cb67efb8c8a4e9ff3a52c68b55f6e095d1392", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "623a11e288fce5cf37e9f6f52f4ce6bb67b0f8b22a5093e5f482e45a3518c13d", "generation_seconds_batch": 24.829320911085233, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0038645267486572266, "target_z": -0.5273437557798095, "ce": 6.125868320465088, "genuineness": 3.9550557136535645, "identity": 0.25854864716529846, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/25f7dd4083be03f05c73a9806a6cb67efb8c8a4e9ff3a52c68b55f6e095d1392.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "262a0d8ae7e987db5d748a459092add2898b4e05ac0f139e9950ff76314b4040", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "957ed8a6a1111f5c4c3d624bea935c0d0e3dc88cc790d67946a867b7d5c7f0f3", "generation_seconds_batch": 24.96092503296677, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6236945986747742, "target_z": 0.6562500248976408, "ce": 6.1894097328186035, "genuineness": 4.356721878051758, "identity": 0.5650544166564941, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/262a0d8ae7e987db5d748a459092add2898b4e05ac0f139e9950ff76314b4040.mp3" }, { "id": "heldout-en-1", "language": "en", "sentences": [ "The keys are on the kitchen table.", "I checked the drawer... then looked beside the window.", "Take them with you when you leave." ], "text": "The keys are on the kitchen table. I checked the drawer... then looked beside the window. Take them with you when you leave.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-1", "key": "26d12995a5b024b42cbe037fdee48ffb67b79df656c09f059592185aa881aca9", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The keys are on the kitchen table.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The keys are on the kitchen table.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Intense Malevolence Malice. 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": "d86be4688ee5d3baaed36adc18a637c74731530a83e9eb598135db6ce8be2789", "generation_seconds_batch": 22.634793738950975, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.008409321308135986, "target_z": -0.5507812282145643, "ce": 6.435632705688477, "genuineness": 4.20609712600708, "identity": 0.248957097530365, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you leave.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/26d12995a5b024b42cbe037fdee48ffb67b79df656c09f059592185aa881aca9.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": "26ec5296003896dc14f53970e6e7f34f60b060de353ab3b97510ecc338b0722a", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e24303de54d57c487f64ce99be5eae35cb7f29f70156e8b420d731675638893d", "generation_seconds_batch": 22.95551992196124, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.007955819368362427, "target_z": -0.5195312459985935, "ce": 6.252951145172119, "genuineness": 4.250029563903809, "identity": 0.6803587079048157, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/26ec5296003896dc14f53970e6e7f34f60b060de353ab3b97510ecc338b0722a.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "270005d31c2083b420c59a912588b5fd5fe305672d7fa535ae7cf85af6421f86", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "3706619686ee518ad79868fc213d2f6bdb0aa1a56d01bdabf642314f67d75145", "generation_seconds_batch": 25.650032219011337, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.031097412109375, "target_z": 3.3437498612845724, "ce": 6.260893821716309, "genuineness": 4.306513786315918, "identity": 0.44103139638900757, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/270005d31c2083b420c59a912588b5fd5fe305672d7fa535ae7cf85af6421f86.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": "2716ffe578a931bbb49a8a5906ce1113ed5c932357f3ddbc3a1d1e5fbceb8f5b", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Your message arrived after I had left.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "51692b9ce20c8c234a49ea86ca61120cec1acef5693f9482a60e9cc9966f3a13", "generation_seconds_batch": 17.246038323035464, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.059096917510032654, "target_z": -0.4218749875511796, "ce": 6.324435234069824, "genuineness": 3.8922953605651855, "identity": 0.3672259449958801, "hypothesis": "Your message arrived after I had left. I read it twice, before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/2716ffe578a931bbb49a8a5906ce1113ed5c932357f3ddbc3a1d1e5fbceb8f5b.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "27d6722d7452d47c7e76143f66977a7e08f4b264062d47c565a8bce2ecfbde35", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "68845e36f4b8d97120cf677260f67636907fe2cd68d11fa26d7f1a0f432a53e4", "generation_seconds_batch": 27.233545611030422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.0493222177028656, "target_z": -0.6289062691178313, "ce": 6.372091293334961, "genuineness": 4.243753433227539, "identity": 0.4372543394565582, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/27d6722d7452d47c7e76143f66977a7e08f4b264062d47c565a8bce2ecfbde35.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "27de68f0d7d389e45f6c115887b88bd7ae7926ab1a934b6ff0d6b95b04f427ee", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "882d2bd48bbca2b2f75467391979d48a24ee4fb35dc2adb439d4c0945ba59b0b", "generation_seconds_batch": 28.543067724909633, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.05136784911155701, "target_z": -0.6328124955539927, "ce": 6.403861999511719, "genuineness": 4.319066047668457, "identity": 0.2503698468208313, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/27de68f0d7d389e45f6c115887b88bd7ae7926ab1a934b6ff0d6b95b04f427ee.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": "285aa7c2505d83a60da493c292a2ba6b4425aa74dbe996b0c1e5da9f75d4270c", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "117785efd8bdfa232f83c9dca7a17119074cb83040dc91e020c3e3648214dbda", "generation_seconds_batch": 23.27837068203371, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.016138404607772827, "target_z": -0.5039062264361613, "ce": 6.260893821716309, "genuineness": 4.325342178344727, "identity": 0.6426747441291809, "hypothesis": "De weg gaat linksaf naar de brug. Loop verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/285aa7c2505d83a60da493c292a2ba6b4425aa74dbe996b0c1e5da9f75d4270c.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "288a719932656b118061f7d4e0ea1ef56ff5431613e7eaa5af53a5f9e2de4cd5", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "f02cb04a524cfcc21da8c694e9219669d9d320b3142465823026d6542687bee4", "generation_seconds_batch": 27.423376059974544, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0018188953399658203, "target_z": -0.5312499822159709, "ce": 5.998785495758057, "genuineness": 4.174716949462891, "identity": 0.5441896915435791, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/288a719932656b118061f7d4e0ea1ef56ff5431613e7eaa5af53a5f9e2de4cd5.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "28c169c11b5bfc8bb7605d5372fcc6ca93ac9376dafe040bbf2b74a90b446851", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "99efca5b568e6d520e887cbb82d68efe7c26dff9d9e8fbdfb68da1db7b543691", "generation_seconds_batch": 27.423376059974544, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.010001450777053833, "target_z": -0.5156250195624321, "ce": 6.205295085906982, "genuineness": 4.281409740447998, "identity": 0.6081657409667969, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/28c169c11b5bfc8bb7605d5372fcc6ca93ac9376dafe040bbf2b74a90b446851.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": "28c171c48b11014a1bc527abbcc5a16d70bb3ac2aa926960b96f6d48a1bf9829", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The road turns left behind the bridge.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c9198f8ea0c3bcb85614c806dc74bffffddd18a28493f65a7e08b4cb70efe08a", "generation_seconds_batch": 20.78920121199917, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3273319602012634, "target_z": 0.09033205542516463, "ce": 6.340320587158203, "genuineness": 4.005263805389404, "identity": 0.18445734679698944, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/28c171c48b11014a1bc527abbcc5a16d70bb3ac2aa926960b96f6d48a1bf9829.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "28e652583298d22e772a370302131a9b3db56a391f368853d0eb66567f0b3d47", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ebad2d8abce2ed4a98ca160e4bae39fd12e52ced88911dfb3854a3f9c205ed27", "generation_seconds_batch": 25.495932615944184, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.66288161277771, "target_z": 2.6406250053352087, "ce": 6.117925643920898, "genuineness": 3.823258876800537, "identity": 0.4519839584827423, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude graam. We kunnen er samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/28e652583298d22e772a370302131a9b3db56a391f368853d0eb66567f0b3d47.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "291a60a26c9fc0aa71464da4cb839cf247e150c373e86fa5641ae32b87c95d04", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c56048d587ef2aa3a34f2048d026a2d38d06eb2a480e58e59e23ffa60e1aaf18", "generation_seconds_batch": 18.91052573500201, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5500514507293701, "target_z": 0.5156250764713254, "ce": 6.252951145172119, "genuineness": 4.243753433227539, "identity": 0.4786244332790375, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/291a60a26c9fc0aa71464da4cb839cf247e150c373e86fa5641ae32b87c95d04.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": "29823e491d39612271d163f9a4e31f706725ae5d3195ce63f24dcd353fc8b7f2", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "74baac20486c4622fbdf1c0661c770f47ea0ada5aaf3a2aec16d6752737402ca", "generation_seconds_batch": 17.12099462095648, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03454919159412384, "target_z": -0.4687499893295825, "ce": 6.229123115539551, "genuineness": 4.532451152801514, "identity": 0.5261460542678833, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/29823e491d39612271d163f9a4e31f706725ae5d3195ce63f24dcd353fc8b7f2.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "29f6478eaa5ca964ba30565cfc99799f288582bf553071d9c42c5543164a1dfc", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "ed61ad01d20a82cacd91da20f7c39791b0797926259a750a71b3c4c9195742a7", "generation_seconds_batch": 17.868277807021514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8183505535125732, "target_z": 2.9375000355680583, "ce": 6.1894097328186035, "genuineness": 4.218649387359619, "identity": 0.4211944341659546, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/29f6478eaa5ca964ba30565cfc99799f288582bf553071d9c42c5543164a1dfc.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "2afca5f50f0808f240b370b7412ec336258a91037627defa40eb0e8ea00dcd55", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "b64cc13bedd90f8e6dc7df774845fced7c54f90a63726ea91d14fe07d4a82cdd", "generation_seconds_batch": 27.501419786014594, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00022676587104797363, "target_z": -0.5351562655610255, "ce": 6.070269584655762, "genuineness": 4.243753433227539, "identity": 0.604188084602356, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/2afca5f50f0808f240b370b7412ec336258a91037627defa40eb0e8ea00dcd55.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "2b6c9002527975dc4291257cc384fbb4143522adf44bbc0fb33183f70e3a0153", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "83fb90a4e00d64e76d033882f06839ff80adda39fdbb95f9dd8d21a061138807", "generation_seconds_batch": 26.260819707997143, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03045789897441864, "target_z": -0.47656249911079857, "ce": 6.1894097328186035, "genuineness": 4.306513786315918, "identity": 0.3713392913341522, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/2b6c9002527975dc4291257cc384fbb4143522adf44bbc0fb33183f70e3a0153.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": "2ba0bb83a6ce0580db9820a279cb1d39218a44b309f7c98095f16dc6694edabd", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d78d07591fdf52f8136c00e688cb273db1a885c59556fc8797939e75e7e22881", "generation_seconds_batch": 25.858792957034893, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8118938207626343, "target_z": 1.0156250195624321, "ce": 6.41180419921875, "genuineness": 4.595211505889893, "identity": 0.6085202693939209, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/2ba0bb83a6ce0580db9820a279cb1d39218a44b309f7c98095f16dc6694edabd.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "2bea7564d00f82c1d0bb5ea2269cf589f178fe2fd340e4fc6bbf4cc7e3258879", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "ee204b4831d67ff90f27f321d208845a90a7b54b9c436d7d5d4a01bee4b2dc23", "generation_seconds_batch": 27.606042865896598, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0778274536132812, "target_z": 1.5234374155258614, "ce": 6.133810997009277, "genuineness": 4.193545341491699, "identity": 0.28216108679771423, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/2bea7564d00f82c1d0bb5ea2269cf589f178fe2fd340e4fc6bbf4cc7e3258879.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "2bf45426b9571e9c0132bd561754159e305f84d3cfa452c70a846c550f92e820", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "a5709e248994bc1325664f288e30531b037c939e5786bbdfb1740545f8b6543d", "generation_seconds_batch": 19.062127122073434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.025343775749206543, "target_z": -0.48632815056454193, "ce": 5.998785495758057, "genuineness": 4.1998209953308105, "identity": 0.6369661092758179, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/2bf45426b9571e9c0132bd561754159e305f84d3cfa452c70a846c550f92e820.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "2c78b777181e0ae7387285135901b9a89495713f484ed594c344748ce15f8174", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c0c3c8367f5b6a826fd8ccf50d1b393b954969be35bfde28ae1c1a0c54d49293", "generation_seconds_batch": 19.91579747700598, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1432881355285645, "target_z": 1.6484375720253182, "ce": 6.23706579208374, "genuineness": 4.243753433227539, "identity": 0.3329504132270813, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/2c78b777181e0ae7387285135901b9a89495713f484ed594c344748ce15f8174.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "2dbf6922c974d9a4cb0d0ef2aa6f6001e47652fe2ff7aba442213ed03e4ecebd", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "ec43de623c481cb29191b106a5d936972d55803c2760b287c9a40c647c775995", "generation_seconds_batch": 27.131395080010407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.05545914173126221, "target_z": -0.6406250053352087, "ce": 6.086154937744141, "genuineness": 4.331617832183838, "identity": 0.7186555862426758, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/2dbf6922c974d9a4cb0d0ef2aa6f6001e47652fe2ff7aba442213ed03e4ecebd.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "2de36ad2c62e8619e36be48a49f452c02d4a2263b23494b3e564bfb140ee4011", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3027119b0f37e2d525253407a647eb479efc119da37a9e3d11789d0ced145abe", "generation_seconds_batch": 27.129553152946755, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.153836250305176, "target_z": 3.57812526853884, "ce": 5.974957466125488, "genuineness": 4.168440818786621, "identity": 0.6350620985031128, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/2de36ad2c62e8619e36be48a49f452c02d4a2263b23494b3e564bfb140ee4011.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "2de6494f2e4f54deb42cc88c5aca1f1babde3b6222e9834e971dbf925bbb99e9", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "6eca0653588440ea11aa353d1fe2baa7973e30279008570a354e20ae6fa5d379", "generation_seconds_batch": 18.38138328201603, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.055005624890327454, "target_z": -0.42968749733239564, "ce": 6.284721851348877, "genuineness": 3.9927117824554443, "identity": 0.6388759016990662, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/2de6494f2e4f54deb42cc88c5aca1f1babde3b6222e9834e971dbf925bbb99e9.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": "2df1683b10d001e41f74eb7c3528c5466cb3ef4adb929469a8f74cedb057e3b6", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The keys are on the kitchen table.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0d6c40a556e0deab17973c63a474a8146118c7c374a0bee3a33ab8bddcaf8462", "generation_seconds_batch": 17.22400686191395, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.039094001054763794, "target_z": -0.6093750231192379, "ce": 6.3641486167907715, "genuineness": 4.093128204345703, "identity": 0.36593714356422424, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you leave.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/2df1683b10d001e41f74eb7c3528c5466cb3ef4adb929469a8f74cedb057e3b6.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "2e298fee7021adbfddaeadc639d8dc8896eb7aa293d2d16140228a43ee9257ee", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "f8e91cdff29b2ec3344e7353c18850ff78ec6a9655e13fc72970661daf0ea379", "generation_seconds_batch": 19.430608466966078, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1266031563282013, "target_z": -0.2929687468877949, "ce": 6.38003396987915, "genuineness": 3.9425034523010254, "identity": 0.36147019267082214, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/2e298fee7021adbfddaeadc639d8dc8896eb7aa293d2d16140228a43ee9257ee.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "2f7b811ddd67edc141196ad5ee6c08eeac62ca167b61cd439e5ec056f29b8134", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "7806ba3997b2496ee8966c9fa2fe71d81ece6c9b3cfa059b7ce83e054cccd642", "generation_seconds_batch": 21.487986081046984, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05398279428482056, "target_z": -0.43164063900492294, "ce": 6.213237762451172, "genuineness": 4.356721878051758, "identity": 0.6403074264526367, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/2f7b811ddd67edc141196ad5ee6c08eeac62ca167b61cd439e5ec056f29b8134.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "2f7e071fc677b79f617e14a270d53a523868c7695a2102cf622ca620e5cb8bb1", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "6fd13d579e118cdebf38d2399da81a3a82c722ab093db8c0aa77717198379436", "generation_seconds_batch": 27.501419786014594, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02125251293182373, "target_z": -0.49414060343686467, "ce": 6.070269584655762, "genuineness": 4.042920112609863, "identity": 0.6993745565414429, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/2f7e071fc677b79f617e14a270d53a523868c7695a2102cf622ca620e5cb8bb1.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "2fc1b537c731fdd0b3f8234beed00a1fc3b0397b1b4a6b6e37c5e55a58231a4b", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "9b89e222f54dfc775e0fe273edb76867cc5b963668cec09f561a9e306b550e72", "generation_seconds_batch": 23.3032032570336, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.027389436960220337, "target_z": -0.48242186721948727, "ce": 5.824047088623047, "genuineness": 3.7793264389038086, "identity": 0.5881825685501099, "hypothesis": "La puerta sigue abierto tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.2, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/2fc1b537c731fdd0b3f8234beed00a1fc3b0397b1b4a6b6e37c5e55a58231a4b.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "3092c69678745f5c40945af650f6afb7de90a04c5ee3a38bfdf2302624ef6ecd", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "c1fbb82be50d0921e3031ae71258fd6eb023168798cb48318b0108af3d66f0bf", "generation_seconds_batch": 24.48615619703196, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.012047111988067627, "target_z": -0.5117187362173774, "ce": 6.141753673553467, "genuineness": 4.325342178344727, "identity": 0.5250654220581055, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/3092c69678745f5c40945af650f6afb7de90a04c5ee3a38bfdf2302624ef6ecd.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": "3096b1d11ce255190622043cb6c44c308f03613e79a1a2c4e2e7a4df734a8a28", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Your message arrived after I had left.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7c696cec90f3dcd081ac9e83ddb01701855e13c6e79f79848ffb2e50974622e1", "generation_seconds_batch": 17.246038323035464, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.018637537956237793, "target_z": -0.5703124742131577, "ce": 6.475345611572266, "genuineness": 4.080576419830322, "identity": 0.38857221603393555, "hypothesis": "Your message arrived after I had left. I read it twice, before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/3096b1d11ce255190622043cb6c44c308f03613e79a1a2c4e2e7a4df734a8a28.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "3099bbbb0a5f72ce9d52a46a860678c00bc60a89b29a086999c5d838e05fbb82", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "9c34424d843c139673ba82d9a6f756c9f2d3dc8aa693a3756e86bd08e27f3948", "generation_seconds_batch": 28.599422290106304, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10614672303199768, "target_z": -0.33203123888498176, "ce": 6.054384231567383, "genuineness": 3.8044304847717285, "identity": 0.3441898822784424, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/3099bbbb0a5f72ce9d52a46a860678c00bc60a89b29a086999c5d838e05fbb82.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "31980dd18a220f20dda980f8b518d666b6ca5b1347abd259031f4818bd88975c", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "b2f9bf6879dc3d26059808e4832675294e528e06b0ce2cdc3dc224962e933d7e", "generation_seconds_batch": 22.26370304694865, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08159899711608887, "target_z": -0.3789062406633847, "ce": 6.221180438995361, "genuineness": 4.457139015197754, "identity": 0.4040134847164154, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/31980dd18a220f20dda980f8b518d666b6ca5b1347abd259031f4818bd88975c.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "31fba6fd43001b3a0da2d07c6fc12e426c387c1446b0d458f3df1ca2e1eeb08d", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "457342c9e5390b2a22d7195f1afac92fce3473ed781f47edba108337fcfa58c6", "generation_seconds_batch": 28.83332093397621, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.0411396324634552, "target_z": -0.6132812495553993, "ce": 6.221180438995361, "genuineness": 4.507347106933594, "identity": 0.431148886680603, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/31fba6fd43001b3a0da2d07c6fc12e426c387c1446b0d458f3df1ca2e1eeb08d.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "325205a5931efbb4efcc48bd2ae7d7ab9d38504e67ed652dcd2d95ecc4dbc78b", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "9eefeb2f1be0936b68d441d13cbe68ea35cf484396ddd3dbf9ec76b089f35bf0", "generation_seconds_batch": 27.862253987928852, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.05750477313995361, "target_z": -0.6445312317713701, "ce": 6.268836498260498, "genuineness": 4.049196243286133, "identity": 0.5047867298126221, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/325205a5931efbb4efcc48bd2ae7d7ab9d38504e67ed652dcd2d95ecc4dbc78b.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": "33f06e46b435ac30ed295e8474673b5916090c443a09d70c63e78b42d2756695", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The keys are on the kitchen table.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eb5782a63603570f1380fa263bb23ca8be66927949400da2a3bcf289af76e9da", "generation_seconds_batch": 18.254465957987122, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.026820123195648193, "target_z": -0.5859374937755898, "ce": 6.443574905395508, "genuineness": 4.256305694580078, "identity": 0.42963936924934387, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you leave.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/33f06e46b435ac30ed295e8474673b5916090c443a09d70c63e78b42d2756695.mp3" }, { "id": "heldout-nl-2", "language": "nl", "sentences": [ "De weg gaat linksaf na de brug.", "Loop verder... tot aan het kleine houten hek.", "Ik wacht aan de andere kant." ], "text": "De weg gaat linksaf na de brug. Loop verder... tot aan het kleine houten hek. Ik wacht aan de andere kant.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-2", "key": "33f63538ac2c49676cc53462aa2024b47e58cb29b8690774a6580785ecb0e60e", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Intense Malevolence Malice. 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": "8fcdf129ade63719025dbabd5a230e43b13be31b1c601411ed05cda9d907ebf6", "generation_seconds_batch": 18.9423877610825, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0018188953399658203, "target_z": -0.5312499822159709, "ce": 6.133810997009277, "genuineness": 4.231201171875, "identity": 0.6565718650817871, "hypothesis": "De weg gaat linksaf naar de brug. Loop verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/33f63538ac2c49676cc53462aa2024b47e58cb29b8690774a6580785ecb0e60e.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "33fc52394e71205f5c3e7f4c5f4eee3137720fe764f9818e2efac01ad61f229b", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "5955d53fe92380140307fa8d89ce28a834d10e2e96366a7c74571f7581d2ca1a", "generation_seconds_batch": 26.761937272036448, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01045498251914978, "target_z": -0.5546875115596189, "ce": 6.205295085906982, "genuineness": 4.557555198669434, "identity": 0.718077540397644, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/33fc52394e71205f5c3e7f4c5f4eee3137720fe764f9818e2efac01ad61f229b.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "340d34f440200e6dfc7e29c3e2fec76a50fc21a4eb804d9c4cedb23d479baed0", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "f6ee0b60b3fb65371ae6b74a787874368d4de8820fa8edb4279d46400e5f7933", "generation_seconds_batch": 25.769147370010614, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.04727655649185181, "target_z": -0.6249999857727767, "ce": 6.324435234069824, "genuineness": 4.030367851257324, "identity": 0.32214125990867615, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/340d34f440200e6dfc7e29c3e2fec76a50fc21a4eb804d9c4cedb23d479baed0.mp3" }, { "id": "heldout-en-2", "language": "en", "sentences": [ "The road turns left behind the bridge.", "Keep going... until you reach the small wooden gate.", "I will meet you on the other side." ], "text": "The road turns left behind the bridge. Keep going... until you reach the small wooden gate. I will meet you on the other side.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-2", "key": "347b084eb1c87968a3fc14b855043a2de140e81cf0874b7d65525285bfb31414", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The road turns left behind the bridge.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The road turns left behind the bridge.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Intense Malevolence Malice. 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": "f638d14fd47ff29dde65ff92b5bf472c9c35946963ca522c2737a5ee0cce7e59", "generation_seconds_batch": 17.580194295034744, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.016591906547546387, "target_z": -0.5664062477769963, "ce": 6.523001670837402, "genuineness": 4.13078498840332, "identity": 0.2661299705505371, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/347b084eb1c87968a3fc14b855043a2de140e81cf0874b7d65525285bfb31414.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "347c34ad15faaeb0a8614e42594ff9f7db83c97e8aee420cac6db43547603786", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "e1f12b7736239c036ba0cad6d93e21e70d60be53de467edbb79e412a85d71131", "generation_seconds_batch": 26.802801141981035, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03352636098861694, "target_z": -0.47070313100210986, "ce": 6.197352409362793, "genuineness": 3.9927117824554443, "identity": 0.2781146168708801, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/347c34ad15faaeb0a8614e42594ff9f7db83c97e8aee420cac6db43547603786.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "34b780530282be6d32b7efc6ee38d5c36ebc636bf20cc47655d1f99abe50ccfc", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": 127, "seconds_generated": 10.16, "mp3_sha256": "71e1aa78a1f17edb30ec55de739e43ea87b14715d0d412b8476908582e4896a8", "generation_seconds_batch": 18.499824290978722, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06318819522857666, "target_z": -0.4140625062244102, "ce": 6.356205940246582, "genuineness": 4.055472373962402, "identity": 0.4300994575023651, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/34b780530282be6d32b7efc6ee38d5c36ebc636bf20cc47655d1f99abe50ccfc.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": "3512e15f048a014dc8bd3e395741cd3668a9634416116ad114181918141de0eb", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Someone moved the little blue chair.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d2b19699a55f6ac2b9f27edde538907f91b4e4ce7d5357016afe4896072d4c64", "generation_seconds_batch": 27.68157036800403, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0059101879596710205, "target_z": -0.5234374724347548, "ce": 6.3641486167907715, "genuineness": 4.406930446624756, "identity": 0.37097078561782837, "hypothesis": "Some one moved the little blue chair. It used to stand here, beside the tall bookshelf. We can put it back after lunch.", "wer": 0.09090909090909091, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/3512e15f048a014dc8bd3e395741cd3668a9634416116ad114181918141de0eb.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "35b32a6d8651deaac318370b20a3875c9e82f019cb1ce3e8843096b2745806a5", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "2c98f9a6e8adf1c6578de33b07e0bcb0a07a6a6acb59a1c017b2ffe51f353b69", "generation_seconds_batch": 26.544314936967567, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03045789897441864, "target_z": -0.47656249911079857, "ce": 6.252951145172119, "genuineness": 4.124508857727051, "identity": 0.07885327935218811, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/35b32a6d8651deaac318370b20a3875c9e82f019cb1ce3e8843096b2745806a5.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "35d31113a324db958f7fcb54a159ac82d90f88dd661318fe79d4289bfadfe214", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "3445da328fb43723e9c08545ba6819dfbad7f5ab9b7ba835aed5fe5e00c0d1da", "generation_seconds_batch": 26.189175114035606, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01045498251914978, "target_z": -0.5546875115596189, "ce": 6.316492557525635, "genuineness": 4.519899368286133, "identity": 0.595215380191803, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/35d31113a324db958f7fcb54a159ac82d90f88dd661318fe79d4289bfadfe214.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "35e2318e783f11246e875d8d6dfe692f083022897e97bd19e7b972b0432eaa62", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "508f6f496c4c2afa7020fbe577aa93d983279dd82dec54cda46f7174dcf8822e", "generation_seconds_batch": 20.72498626401648, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.20740607380867004, "target_z": -0.13867189211712808, "ce": 5.919358730316162, "genuineness": 3.9801597595214844, "identity": 0.2591343820095062, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/35e2318e783f11246e875d8d6dfe692f083022897e97bd19e7b972b0432eaa62.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "35f11f40b91b503176328edda71a1c6ee7fe84af1c6d029c97c219311f9ce1e6", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "e9540368d9261f567cd2e99ed3e5b219d82bf4a318d62de8960615c64b4bc92c", "generation_seconds_batch": 19.0847464739345, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01454627513885498, "target_z": -0.562500021340835, "ce": 5.808161735534668, "genuineness": 3.9173994064331055, "identity": 0.3355550765991211, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/35f11f40b91b503176328edda71a1c6ee7fe84af1c6d029c97c219311f9ce1e6.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "367e3071b3d6d7b0dec59b5cccd27f951d1fa50035c06cfb791aaf92ee0780aa", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "f5d95bc44e495a42e9fb3eb731ab30933e2bd55017f68a00bc494836d4ae568a", "generation_seconds_batch": 23.85318032791838, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02841225266456604, "target_z": -0.48046875400140654, "ce": 6.038498878479004, "genuineness": 4.3127899169921875, "identity": 0.5901399850845337, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/367e3071b3d6d7b0dec59b5cccd27f951d1fa50035c06cfb791aaf92ee0780aa.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": "37261896945348f35075c880a4943eaa6c0ca3648a2ec828e221d9d24714b67b", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f36bc38c160e5d9c8f39ea314d9c2848a8efa2b3f2de82b23022ed74dfdd40ca", "generation_seconds_batch": 18.721526676090434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.022728830575942993, "target_z": -0.5781249839943737, "ce": 6.268836498260498, "genuineness": 4.306513786315918, "identity": 0.5612773895263672, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/37261896945348f35075c880a4943eaa6c0ca3648a2ec828e221d9d24714b67b.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "3744bdd999f55aeae8076cf73beca487b1a43bd9661d9258aa162e7ac8506880", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f0afd0508782262efff2aaf883d2fd4362eb3bb6cc0cc6b731555cb08a477058", "generation_seconds_batch": 29.034140834002756, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.010001450777053833, "target_z": -0.5156250195624321, "ce": 6.268836498260498, "genuineness": 4.030367851257324, "identity": 0.21607735753059387, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/3744bdd999f55aeae8076cf73beca487b1a43bd9661d9258aa162e7ac8506880.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "3771fefed616bd21de8190cd7a440c3cfd8b2239901722af0f2569e5a6b639e7", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "015601c96940b41ee356c2b448bc8bc456e56360fe3d601e4faa8cc449870f93", "generation_seconds_batch": 22.662760969018564, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2650037407875061, "target_z": -0.028686521117239946, "ce": 6.133810997009277, "genuineness": 3.8922953605651855, "identity": 0.17345359921455383, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/3771fefed616bd21de8190cd7a440c3cfd8b2239901722af0f2569e5a6b639e7.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "37dce0afc7e884b07b4f5ede3565deddb88f234f9dde2f54c74bb9b89fed5919", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "f0057f5f7302b488a7b173f0447d02e92593a3601450afba260150fe3d962bb4", "generation_seconds_batch": 22.526967811980285, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5602796077728271, "target_z": 0.5351562086521322, "ce": 6.165581703186035, "genuineness": 3.6789097785949707, "identity": 0.20734700560569763, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/37dce0afc7e884b07b4f5ede3565deddb88f234f9dde2f54c74bb9b89fed5919.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "37f2fbe3d394433e577b19ec30b1b5c787fbdefc5f541900a16c54fd751b5dea", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "24a0dafc6c8dcfd2754b63b2c5c25147b1f8e1aa3cf7faa6ae726252ec810378", "generation_seconds_batch": 24.19068988203071, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.05750477313995361, "target_z": -0.6445312317713701, "ce": 6.403861999511719, "genuineness": 4.432034492492676, "identity": 0.25394564867019653, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/37f2fbe3d394433e577b19ec30b1b5c787fbdefc5f541900a16c54fd751b5dea.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "37fde071f1bd4179c08274f00c6dfb0f4298897545d2c649c7e02c31d1cd55d7", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "1f74c2c0a2c10585961f081ce07dc3aa3fc2e33a7e46a722adb16b92b417b8f1", "generation_seconds_batch": 20.82998289493844, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.016591906547546387, "target_z": -0.5664062477769963, "ce": 6.006728172302246, "genuineness": 4.055472373962402, "identity": 0.6005053520202637, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/37fde071f1bd4179c08274f00c6dfb0f4298897545d2c649c7e02c31d1cd55d7.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": "38250daf018b7e6522dff9e2f37cc93eda7a893004315950855567063bfddb3e", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "153a8a247df13818d47ffa53f4bb9bbeb003601bf888f916f2076af4b964185c", "generation_seconds_batch": 22.4599160089856, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02841225266456604, "target_z": -0.48046875400140654, "ce": 6.213237762451172, "genuineness": 4.344170093536377, "identity": 0.5879613161087036, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/38250daf018b7e6522dff9e2f37cc93eda7a893004315950855567063bfddb3e.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "3893968d3cca467658e4c68d3fa7d2af58f2487b37b964e661600747beb85c7b", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "40cba1f4686f2fe7d9e52cd82dae9fe99e4cb3bf1ab61cc1b498225e0c72f0f0", "generation_seconds_batch": 27.46585072297603, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02022966742515564, "target_z": -0.49609377356383866, "ce": 6.109982967376709, "genuineness": 4.356721878051758, "identity": 0.6866892576217651, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/3893968d3cca467658e4c68d3fa7d2af58f2487b37b964e661600747beb85c7b.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": "39343a5ffe21555c7543dfc73b7eff2612c55f459ecb5f834dbf75cd8fca6bab", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ed49deb162756a0ae98df3f35335a7fbf1c148c135e48d98e9b9a6f81f6c222f", "generation_seconds_batch": 17.175388882984407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.004318058490753174, "target_z": -0.5429687753422415, "ce": 6.221180438995361, "genuineness": 4.15588903427124, "identity": 0.5507808923721313, "hypothesis": "Las llaves están en la mesa de la cocina, miré en el cajón, y luego, junto a la ventana, llévatelas cuando te vayas.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/39343a5ffe21555c7543dfc73b7eff2612c55f459ecb5f834dbf75cd8fca6bab.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": "3a3d850327d71b333f274bcc8340dd95fa2d07daf64f4be875d33196a2bdde3a", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The road turns left behind the bridge.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4686238e4395e5c987f8045a6b6ad49467cc56846e8a7fe4e8853896faeedba8", "generation_seconds_batch": 27.128824665094726, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.012500613927841187, "target_z": -0.5585937379957803, "ce": 6.41180419921875, "genuineness": 3.9048471450805664, "identity": 0.4836575388908386, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/3a3d850327d71b333f274bcc8340dd95fa2d07daf64f4be875d33196a2bdde3a.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "3a888752b3b44b029c65d74c78d6bdc32df49b303672fa9befb6bc3f1c085769", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "2424118bcad972da6c70064d956a5af2c59bec1996369ead21140f15f82d7bcd", "generation_seconds_batch": 27.271086927037686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.024774491786956787, "target_z": -0.5820312673394284, "ce": 6.1020402908325195, "genuineness": 4.080576419830322, "identity": 0.5807143449783325, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/3a888752b3b44b029c65d74c78d6bdc32df49b303672fa9befb6bc3f1c085769.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "3b3a76085e0e7ca5b473773817795cb25ac00219722fac20fc673257414e4f69", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "2fc80343d5b1686fe2ace1881d05b5b26a0c6327589e51bc4905cb7fbed0ded5", "generation_seconds_batch": 27.233545611030422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.043185263872146606, "target_z": -0.6171874759915607, "ce": 6.23706579208374, "genuineness": 4.18099308013916, "identity": 0.4327402412891388, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/3b3a76085e0e7ca5b473773817795cb25ac00219722fac20fc673257414e4f69.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "3b991ae6a0669697c74d6c659971d319b8ab01bfed64fcfcfead1efc0d20c00b", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "45187663f2dc7c249ce9a515cce36c961e5e9b0a0c3eb57a4da2955164c62552", "generation_seconds_batch": 19.622488979948685, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.06364172697067261, "target_z": -0.6562500248976408, "ce": 6.308549880981445, "genuineness": 4.080576419830322, "identity": 0.42744219303131104, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/3b991ae6a0669697c74d6c659971d319b8ab01bfed64fcfcfead1efc0d20c00b.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "3bc7d599dbec37165a875f959931683438362441e1f38588114e34b7e9ac23ec", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "03f43cd3950ce0adf34fac4f2981102a6990c03c4702fa86353c21c5ae2119bb", "generation_seconds_batch": 26.47648220101837, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.059096917510032654, "target_z": -0.4218749875511796, "ce": 6.427689552307129, "genuineness": 4.061748504638672, "identity": 0.6023284196853638, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/3bc7d599dbec37165a875f959931683438362441e1f38588114e34b7e9ac23ec.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "3bcc2d6f1c4bbfaf36d24f88b23cb13de32a3a4096bab34e5b67ce17f4962f25", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "627c55ae1b43a8ce7e1c88a73452d7dc3d039bcae8564d509fefcbc5a5586baa", "generation_seconds_batch": 31.24500945303589, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.0452309250831604, "target_z": -0.6210937593366153, "ce": 6.38003396987915, "genuineness": 4.049196243286133, "identity": 0.3772686719894409, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/3bcc2d6f1c4bbfaf36d24f88b23cb13de32a3a4096bab34e5b67ce17f4962f25.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "3bcc9af1df4d3ef96bdbae5aefa40139144945a9f521e32812c49dba34c79683", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "04af724ae829ce3119d48776d1057d02d1c3d6ed63d72f9f4c2399a26dd091ce", "generation_seconds_batch": 27.86458622198552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.018184036016464233, "target_z": -0.5, "ce": 6.086154937744141, "genuineness": 4.193545341491699, "identity": 0.104312002658844, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/3bcc9af1df4d3ef96bdbae5aefa40139144945a9f521e32812c49dba34c79683.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "3bf2f52d6689c8e275ae6a602b15c4aeaa6f475dbaf72eb1892d611492ffb6e8", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "56d478301696046e1ba5f981cf96213927a72dcc820e0223d96526e8ae91b20e", "generation_seconds_batch": 28.219675490050577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.010001450777053833, "target_z": -0.5156250195624321, "ce": 6.046441555023193, "genuineness": 4.362998008728027, "identity": 0.4159826636314392, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/3bf2f52d6689c8e275ae6a602b15c4aeaa6f475dbaf72eb1892d611492ffb6e8.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "3c0e3cd4eb20d139e59651cf11977fb09700766b8781f8ae16668cbdd106cd69", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c16143b21791467258e0451c5735c9e8c1f59e7659e36c379b55355120b8ba5d", "generation_seconds_batch": 18.758713988005184, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.024774491786956787, "target_z": -0.5820312673394284, "ce": 5.967014789581299, "genuineness": 4.331617832183838, "identity": 0.6241213083267212, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/3c0e3cd4eb20d139e59651cf11977fb09700766b8781f8ae16668cbdd106cd69.mp3" }, { "id": "heldout-nl-1", "language": "nl", "sentences": [ "De sleutels liggen op de keukentafel.", "Ik keek in de la... en daarna naast het raam.", "Neem ze mee als je weggaat." ], "text": "De sleutels liggen op de keukentafel. Ik keek in de la... en daarna naast het raam. Neem ze mee als je weggaat.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-1", "key": "3c590f9302048ad206a88d3c2440f1028cc9eea1f1662250baf365be21a6ebfa", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Intense Malevolence Malice. 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": "32a6426a79656d1cd6be9929ff7cb26b25923db14b6241915a38afdc50c2b817", "generation_seconds_batch": 20.599867051001638, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.014092743396759033, "target_z": -0.5078125097812161, "ce": 6.23706579208374, "genuineness": 4.369274139404297, "identity": 0.624521017074585, "hypothesis": "De sleutels liggen op de keukentafel. Ik kijk in de la en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/3c590f9302048ad206a88d3c2440f1028cc9eea1f1662250baf365be21a6ebfa.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "3c5a2031c040f3b0307ec37e3605df577ce87d27bd0d983744cdaac19ae55c5c", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1cf817876ffaf44ea2ca9103d69d2ad3865b559d472f4ec43f3a5655cc0dd5bc", "generation_seconds_batch": 26.796472709043883, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.022728830575942993, "target_z": -0.5781249839943737, "ce": 6.109982967376709, "genuineness": 4.218649387359619, "identity": 0.5600243210792542, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/3c5a2031c040f3b0307ec37e3605df577ce87d27bd0d983744cdaac19ae55c5c.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "3ce6ab4fd771e79287d6617a76e25e308dccb8427f1829e395d06afc95b0e5f3", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "c5d8854050095ad53c390a8374dbd65d8a517e5c38d6df8d8879f2a46008a371", "generation_seconds_batch": 26.788008840056136, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2659626305103302, "target_z": -0.026855477475289304, "ce": 6.23706579208374, "genuineness": 3.7542223930358887, "identity": 0.5965868234634399, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/3ce6ab4fd771e79287d6617a76e25e308dccb8427f1829e395d06afc95b0e5f3.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "3db841d9521b3a11bf03adb0adace3bfce6a248322ff8ab50dc4e372fd514b4d", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "0be5a76b4cf12a39e7e45f9d6858d8564ce442cc0520a5926d5fd97b1e953a20", "generation_seconds_batch": 26.848008858039975, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2079174965620041, "target_z": -0.13769530705364108, "ce": 6.229123115539551, "genuineness": 4.18726921081543, "identity": 0.6067160964012146, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/3db841d9521b3a11bf03adb0adace3bfce6a248322ff8ab50dc4e372fd514b4d.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": "3dd8aa2e1682895ca41acd4d02670167ea8d30831e0020c05f4fb8d0a861eba1", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e69913fabf70e4ff8a8b614a7254dfff769177c48b3d70a843582d0adae5e2e0", "generation_seconds_batch": 17.16027115401812, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.026820123195648193, "target_z": -0.5859374937755898, "ce": 6.356205940246582, "genuineness": 3.8797430992126465, "identity": 0.4514163136482239, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade, dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/3dd8aa2e1682895ca41acd4d02670167ea8d30831e0020c05f4fb8d0a861eba1.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "3e0b91103014423524aa542175ccbbcb16165896b3ede7d647395946e8fbf4aa", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "d2a00756ee10ab62a6a0152484d21023239288595ae7b6442431fabfc094212f", "generation_seconds_batch": 21.740140330046415, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.35724949836730957, "target_z": 0.14746095850738444, "ce": 6.1894097328186035, "genuineness": 4.15588903427124, "identity": 0.6487356424331665, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/3e0b91103014423524aa542175ccbbcb16165896b3ede7d647395946e8fbf4aa.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": "3e82ab38ed7c3e5f91594d76a08162437f11c0547064588d909b484c7d27b0fe", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "220f88faff9caaeffce33a235ee7c97285daf25d77a48194e262b65958012055", "generation_seconds_batch": 19.922058008960448, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00227239727973938, "target_z": -0.5390624919971869, "ce": 6.268836498260498, "genuineness": 4.268857479095459, "identity": 0.5046496391296387, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter bis zum kleinen Holztor. Ich warte auf der anderen Seite.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/3e82ab38ed7c3e5f91594d76a08162437f11c0547064588d909b484c7d27b0fe.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "3eaeb95ff471d323acd5496012a2be1ca776de2eadd978553337b669f4615191", "bucket": "emo_Malevolence_Malice", "phase": "wording", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "d51484dda19b82ebcd81de930908135016e55b15969f67f0ba0e2c59c35e0e52", "generation_seconds_batch": 20.101678684004582, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0059101879596710205, "target_z": -0.5234374724347548, "ce": 5.959072113037109, "genuineness": 4.632867813110352, "identity": 0.8153371810913086, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/3eaeb95ff471d323acd5496012a2be1ca776de2eadd978553337b669f4615191.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "3f2c98260ec442b91c784ae406e23d7c912aa6585532e6848877e43abef06d64", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "c0672f5822d8f69de334e4560bc4df99f33265cda6bf76e0dde79bf73f28a305", "generation_seconds_batch": 26.60191045003012, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03557200729846954, "target_z": -0.46679687611150184, "ce": 5.708878040313721, "genuineness": 4.030367851257324, "identity": 0.598189115524292, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/3f2c98260ec442b91c784ae406e23d7c912aa6585532e6848877e43abef06d64.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "3f7fc62527844f934c856afa09aaf57b080942d6644852985ecb8eff37561b8c", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "8c6309318f94fe995dc054efe6bca7fbc6f1e4d0c9a2019431418010610353cd", "generation_seconds_batch": 18.677141251042485, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.010001450777053833, "target_z": -0.5156250195624321, "ce": 6.070269584655762, "genuineness": 4.406930446624756, "identity": 0.594709575176239, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/3f7fc62527844f934c856afa09aaf57b080942d6644852985ecb8eff37561b8c.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "3fabb067bbbab50a1883625f0537be30c4a5240c05b2aa3ab1c9e60213d3255a", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "b23aa7c1f4871516170bcd2a467e6a774d2b8447a7f46a0a3a5743bc00037662", "generation_seconds_batch": 25.346991810016334, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.018184036016464233, "target_z": -0.5, "ce": 5.871703147888184, "genuineness": 4.2249250411987305, "identity": 0.20533406734466553, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/3fabb067bbbab50a1883625f0537be30c4a5240c05b2aa3ab1c9e60213d3255a.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "3ff144e3a92a51ecff59f7d181ee7cb4dd4a2ceb07de6d51d2deb94ae7b30758", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "91806a8ab4e9ac0a7f0d81e4c58aeb627a4b1a7312b4487b09cdf3b44aec393a", "generation_seconds_batch": 26.650074403965846, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00227239727973938, "target_z": -0.5390624919971869, "ce": 6.1894097328186035, "genuineness": 4.293961524963379, "identity": 0.7043479681015015, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/3ff144e3a92a51ecff59f7d181ee7cb4dd4a2ceb07de6d51d2deb94ae7b30758.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "40107616bc48f16cb770c99fec5039d413482bfe3d7b2b72cd347edbee8ae2f6", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "16e15b11996629d08b2ea5929724a44eea9c8b9ca975bc2e62e34d642ce8ab84", "generation_seconds_batch": 27.55493992008269, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00022676587104797363, "target_z": -0.5351562655610255, "ce": 6.252951145172119, "genuineness": 4.545003414154053, "identity": 0.681923508644104, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/40107616bc48f16cb770c99fec5039d413482bfe3d7b2b72cd347edbee8ae2f6.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "403b8d262c1530d51a10d2a1216950b1b120437153e21a0976fbdb9be9da50f7", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "67fdcd8e2d014c3790cdd024010dd5a1f64675c66e4f5d65ddf39a2d1e443b2e", "generation_seconds_batch": 18.758713988005184, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00022676587104797363, "target_z": -0.5351562655610255, "ce": 6.09409761428833, "genuineness": 4.607763767242432, "identity": 0.5544552803039551, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/403b8d262c1530d51a10d2a1216950b1b120437153e21a0976fbdb9be9da50f7.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": "4095de67511b53268e606029cec30c16f7394abce2e774d6b4c7e44d0c028f9a", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I wrote the number on the envelope.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2b44aba262f836a699aa4079536e8d2e1fcdf2cd58cf88aeffc6015563389ea7", "generation_seconds_batch": 17.349562702002004, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08978156745433807, "target_z": -0.36328124955539925, "ce": 6.395919322967529, "genuineness": 3.7291181087493896, "identity": 0.41986072063446045, "hypothesis": "I wrote the number on the envelope. Read it carefully and check the last two digits. Call me when you have a moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/4095de67511b53268e606029cec30c16f7394abce2e774d6b4c7e44d0c028f9a.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "4099163b21c77f5eb01f8a09fd129b46069a198568131b5baf9dd0ed42a82887", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "f638adeb015ae41a930f972e15e169187fef15ff5fd98fb6313ced639a4f7c32", "generation_seconds_batch": 29.87791913608089, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3764914274215698, "target_z": 2.093749889739019, "ce": 6.022613525390625, "genuineness": 4.18726921081543, "identity": 0.39380377531051636, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/4099163b21c77f5eb01f8a09fd129b46069a198568131b5baf9dd0ed42a82887.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "40e2beb065f7ce1a66fa7ba2379c9a80c9b50b395fa958a6c28ebfd39f2fc8f0", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "8570df1889c47b94ea56589f4425eafa8eab2e61c769c07513b5eb854f105420", "generation_seconds_batch": 25.825370609993115, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07546205818653107, "target_z": -0.3906250053352087, "ce": 6.117925643920898, "genuineness": 3.823258876800537, "identity": 0.10882684588432312, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/40e2beb065f7ce1a66fa7ba2379c9a80c9b50b395fa958a6c28ebfd39f2fc8f0.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "40f38591e3cc10c2675cd11a44fea9435415eab2b27193b2aca51bd0aca1e8f5", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bda8a8a39fb89d5c308ff71acdb86ee9210022e61a33fe01d7cefecfc9afbc6d", "generation_seconds_batch": 26.19028902600985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7546157836914062, "target_z": 0.9062499964431942, "ce": 6.205295085906982, "genuineness": 4.149612903594971, "identity": 0.47250327467918396, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/40f38591e3cc10c2675cd11a44fea9435415eab2b27193b2aca51bd0aca1e8f5.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": "4142c316a6a922d0f4b3267e3157b83d5981b4e0e8beb87369d7cc781312ea58", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "30d3aab96c223cef540010a2c270ab9357d72b13c2a5119b35f531c4e359b8de", "generation_seconds_batch": 17.78761451295577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06727948784828186, "target_z": -0.40624999644319415, "ce": 5.772419452667236, "genuineness": 4.042920112609863, "identity": 0.5432559251785278, "hypothesis": "De weg gaat linksaf naar de brug. Loop verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.047619047619047616, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/4142c316a6a922d0f4b3267e3157b83d5981b4e0e8beb87369d7cc781312ea58.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": "41c9ecffe78820062ef73bf70b902220f4ca5e91b00d883c69e5b7ebf7c86e0c", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Someone moved the little blue chair.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3daff6d75f4127bd945d307529fe7f1d5734f22d50f65012c20f2b35d2ccf6d8", "generation_seconds_batch": 17.13475025992375, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00636368989944458, "target_z": -0.5468750017784029, "ce": 6.348263263702393, "genuineness": 4.042920112609863, "identity": 0.3531906008720398, "hypothesis": "Someone moved the little blue chair. It used to stand here beside the tall bookshelf. We can put it back after lunch.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/41c9ecffe78820062ef73bf70b902220f4ca5e91b00d883c69e5b7ebf7c86e0c.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "42d78ea40573be0d25408ea4de19e2c28b611bbeaa68d577b1c8d75a8815103b", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "6ef5bd97786d868c25ec9f6bb9ca7ce8bf242112e07757ce91bd84e758a396d9", "generation_seconds_batch": 26.253196183010004, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05091433227062225, "target_z": -0.43750000711361164, "ce": 5.824047088623047, "genuineness": 3.986435890197754, "identity": 0.4443952739238739, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/42d78ea40573be0d25408ea4de19e2c28b611bbeaa68d577b1c8d75a8815103b.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "4321f9e1d31bd11fb5a9ebe5ed7c8200c86e699c0ff30b635427b3da90928f56", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2cc6ab384f539d1bf80a3c50d97177f5c5c506cf21416e721a2ffd3aa4cb076b", "generation_seconds_batch": 27.440195964998566, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09080438315868378, "target_z": -0.3613281363373186, "ce": 6.41180419921875, "genuineness": 3.9801597595214844, "identity": 0.6804810762405396, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/4321f9e1d31bd11fb5a9ebe5ed7c8200c86e699c0ff30b635427b3da90928f56.mp3" }, { "id": "heldout-nl-2", "language": "nl", "sentences": [ "De weg gaat linksaf na de brug.", "Loop verder... tot aan het kleine houten hek.", "Ik wacht aan de andere kant." ], "text": "De weg gaat linksaf na de brug. Loop verder... tot aan het kleine houten hek. Ik wacht aan de andere kant.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-2", "key": "434f953004ee3838b31e4f505ba0cb9202b34ab37936979fff50edb64a6a996f", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Intense Malevolence Malice. 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": "1ae98f8bbcd667fb5cc0b7e29cde330e62ecacc3e46cd0b5b6ec1c91f73b08b8", "generation_seconds_batch": 18.9423877610825, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.019206851720809937, "target_z": -0.49804688678191933, "ce": 6.197352409362793, "genuineness": 4.042920112609863, "identity": 0.8019531965255737, "hypothesis": "De weg gaat linksaf naar de brug. Loop verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/434f953004ee3838b31e4f505ba0cb9202b34ab37936979fff50edb64a6a996f.mp3" }, { "id": "heldout-es-3", "language": "es", "sentences": [ "Alguien ha movido la pequeña silla azul.", "Estaba aquí... junto a la estantería alta.", "La colocaremos después de comer." ], "text": "Alguien ha movido la pequeña silla azul. Estaba aquí... junto a la estantería alta. La colocaremos después de comer.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-3", "key": "43d771a5c055e2aea1186dc99f4cdef2b4c6bc8c6b2779ad35a29ca11c75a91c", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Intense Malevolence Malice. 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": "d1843d9d8d1e4727a67002ef164aefff2ca941f08d1462ba78f4387555f03da7", "generation_seconds_batch": 19.37162971298676, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.016138404607772827, "target_z": -0.5039062264361613, "ce": 6.165581703186035, "genuineness": 4.356721878051758, "identity": 0.5695948004722595, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí, junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/43d771a5c055e2aea1186dc99f4cdef2b4c6bc8c6b2779ad35a29ca11c75a91c.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "445fc5a2746b8a8c2e222d779b0d41ae239c636628c6dd45d9294fe65780f978", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "c785f710a5cdbfbbe9812ec66191f86418a8b0c3340568ef9d17a702d2881d67", "generation_seconds_batch": 26.28892772900872, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.026366621255874634, "target_z": -0.48437498043756794, "ce": 5.831989288330078, "genuineness": 4.042920112609863, "identity": 0.6072822213172913, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/445fc5a2746b8a8c2e222d779b0d41ae239c636628c6dd45d9294fe65780f978.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "447a156cee16b86d7a9a514e1549501823c3bd80014c4c50f76af3be69364cb5", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "8701748948e9377a4f624338698853440cfdcb8790f3096307b41d61f9081ed5", "generation_seconds_batch": 30.506790620973334, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.023298144340515137, "target_z": -0.49023437700070327, "ce": 5.728734970092773, "genuineness": 3.936227321624756, "identity": 0.15795418620109558, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/447a156cee16b86d7a9a514e1549501823c3bd80014c4c50f76af3be69364cb5.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "449c3e308d0b6135ad18b6cc88fb8c9ff57f8db43218566e8be3f20f43956e36", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "468bc62f6cf183950d0d996081ac6cacd010a417d8e1f9ff00382677cab543bf", "generation_seconds_batch": 27.817798816948198, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0038645267486572266, "target_z": -0.5273437557798095, "ce": 6.046441555023193, "genuineness": 3.8295347690582275, "identity": 0.6548440456390381, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/449c3e308d0b6135ad18b6cc88fb8c9ff57f8db43218566e8be3f20f43956e36.mp3" }, { "id": "heldout-de-4", "language": "de", "sentences": [ "Ich schrieb die Nummer auf den Umschlag.", "Lies sie genau... und prüfe die letzten beiden Ziffern.", "Ruf an, wenn du Zeit hast." ], "text": "Ich schrieb die Nummer auf den Umschlag. Lies sie genau... und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-4", "key": "45f9ec8497c8691f9e18d794cf98f47c522bfab041b36df5accabf5c5c456412", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Intense Malevolence Malice. 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": 129, "seconds_generated": 10.32, "mp3_sha256": "f8e519803015f278cbfe280beb3d5ef5d92bfcef68cdeed94bd1796028f7de38", "generation_seconds_batch": 19.06358708499465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10205543041229248, "target_z": -0.3398437486661978, "ce": 6.3641486167907715, "genuineness": 4.13078498840332, "identity": 0.55164635181427, "hypothesis": "Ich schrieb die Nummer auf den Umschlag. Lies sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/45f9ec8497c8691f9e18d794cf98f47c522bfab041b36df5accabf5c5c456412.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "463c68ca499c0b7e4fe33cd40ff8fd7a0df6ef2251ec644d74d6909a9a211667", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "c1c86cadd064af29961a098ba45aa58260813d1e425b70782155175da391ffb8", "generation_seconds_batch": 27.24187764490489, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.016138404607772827, "target_z": -0.5039062264361613, "ce": 6.1894097328186035, "genuineness": 4.2249250411987305, "identity": 0.5823302865028381, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/463c68ca499c0b7e4fe33cd40ff8fd7a0df6ef2251ec644d74d6909a9a211667.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "464c3a7788f8013d08d019b4f77b21686e26d17a9d3f0511843e5c862c08d4dc", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "2b2ff78ea6cd1cb0bfef70985654b8eb3c7f114c30a2225f52b3f0c2e20fef84", "generation_seconds_batch": 26.97431276203133, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.012047111988067627, "target_z": -0.5117187362173774, "ce": 6.205295085906982, "genuineness": 4.262581825256348, "identity": 0.6859702467918396, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/464c3a7788f8013d08d019b4f77b21686e26d17a9d3f0511843e5c862c08d4dc.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "46c00ac5345c051cf560ad61b9569f28db71fa3485dadd85365e216f7d682074", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a697bd71c2e2d539a6cadaa615b519d6dcd9a3ce0d1a2fc29665550c85f2563b", "generation_seconds_batch": 26.34775253199041, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.020683199167251587, "target_z": -0.5742187575582124, "ce": 6.181467056274414, "genuineness": 4.268857479095459, "identity": 0.5179893970489502, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/46c00ac5345c051cf560ad61b9569f28db71fa3485dadd85365e216f7d682074.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": "46dfdcd1d8a80b037dd427dd0d1019e5564c710e70494d8f590231c54a488ac1", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8d5159b6ae01163a9761a218abf52dce01b95eb96957dcacd448b074366240c6", "generation_seconds_batch": 23.05466240493115, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0059101879596710205, "target_z": -0.5234374724347548, "ce": 6.157639026641846, "genuineness": 4.18726921081543, "identity": 0.6326179504394531, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/46dfdcd1d8a80b037dd427dd0d1019e5564c710e70494d8f590231c54a488ac1.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": "470c068714daa98655e93fd7dc0999368624c50b5448a8e52f4d45dfc8ee85ab", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I wrote the number on the envelope.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2f190aa384482164480ad0ace93ae3045f0a5d1af2ad5cd6d40175cad8608fde", "generation_seconds_batch": 18.92645937402267, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04886868596076965, "target_z": -0.4414062620042197, "ce": 6.2767791748046875, "genuineness": 3.9173994064331055, "identity": 0.46117693185806274, "hypothesis": "I wrote the number on the envelope, read it carefully, and check the last two digits. Call me when you have a moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/470c068714daa98655e93fd7dc0999368624c50b5448a8e52f4d45dfc8ee85ab.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "4771aa5f0a19ccb922bcded4b600bb5cb51f8a1f4f154945bbb398c9b77cc9a2", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "427ba293f21bd248767ab56209d884d1072a4d43c72dfce6bc458f8669d86dc5", "generation_seconds_batch": 21.487986081046984, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05091433227062225, "target_z": -0.43750000711361164, "ce": 6.046441555023193, "genuineness": 4.168440818786621, "identity": 0.6442372798919678, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/4771aa5f0a19ccb922bcded4b600bb5cb51f8a1f4f154945bbb398c9b77cc9a2.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "47bb4425b815749453d857fe4249fd22aae5a6891e0961eb9f8ca7ddaf609967", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "d046c19e0c2db9d46d7915173bb0023c2aa1205fdc55a04a0d5ba3a2f077ce9b", "generation_seconds_batch": 25.678026392008178, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3807743787765503, "target_z": 0.19238279015881338, "ce": 6.022613525390625, "genuineness": 4.13078498840332, "identity": -0.015497570857405663, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/47bb4425b815749453d857fe4249fd22aae5a6891e0961eb9f8ca7ddaf609967.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "47c713ed039666797d5d371a97baaa5cf282ee530ebd784ea0c72360e8e8339f", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "48bc161d6b0dfb4eb848307097662c909205c13c062b290b93e456e9d282de88", "generation_seconds_batch": 27.17870296095498, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.004318058490753174, "target_z": -0.5429687753422415, "ce": 5.760505676269531, "genuineness": 3.886019229888916, "identity": 0.4819694459438324, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/47c713ed039666797d5d371a97baaa5cf282ee530ebd784ea0c72360e8e8339f.mp3" }, { "id": "heldout-fr-3", "language": "fr", "sentences": [ "Quelqu’un a déplacé la petite chaise bleue.", "Elle était ici... près de la grande bibliothèque.", "Nous la remettrons après le déjeuner." ], "text": "Quelqu’un a déplacé la petite chaise bleue. Elle était ici... près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-3", "key": "47d00b404f769e82194cfb39f2be7a6bd3cc5bfaac4f38356b0fada2abdb9f4d", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Intense Malevolence Malice. 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": "9273500608314539d3b66682c430b3fb23d0acd994aaedc778a87d42515ec17f", "generation_seconds_batch": 19.135938877006993, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.026820123195648193, "target_z": -0.5859374937755898, "ce": 6.316492557525635, "genuineness": 4.356721878051758, "identity": 0.7371152639389038, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue. Elle était ici près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/47d00b404f769e82194cfb39f2be7a6bd3cc5bfaac4f38356b0fada2abdb9f4d.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "47ff9f9430589c510deff62740a1e352d4aaa786c4bc5d66c4d35525ddfedfa9", "bucket": "emo_Malevolence_Malice", "phase": "wording", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "e8c386b1295b3ab0f0a0153571c2635df9533fc378f377f65356be052f403290", "generation_seconds_batch": 19.743997538927943, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.022728830575942993, "target_z": -0.5781249839943737, "ce": 6.268836498260498, "genuineness": 3.9048471450805664, "identity": 0.25020164251327515, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/47ff9f9430589c510deff62740a1e352d4aaa786c4bc5d66c4d35525ddfedfa9.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "48eb28c79f298bfe9b1153cf0fdd542c075968e8b6c9b6cd8296aa5a44161d10", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "9da93f247e4eb77d332c5166c4b2b38b75a1624dd6debfd7815a7933a8ab5597", "generation_seconds_batch": 19.78877075400669, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00636368989944458, "target_z": -0.5468750017784029, "ce": 5.967014789581299, "genuineness": 4.331617832183838, "identity": 0.4472401738166809, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg, maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/48eb28c79f298bfe9b1153cf0fdd542c075968e8b6c9b6cd8296aa5a44161d10.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "4900dbb136669906a267938ecde4f7eaa6145c642f76abf26090187c390470bc", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "b1ee3e9a39f72842c6d83881c336e65f269e72b6195f6055c85f3453dd158200", "generation_seconds_batch": 19.966771184001118, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5009559392929077, "target_z": 0.42187495909673295, "ce": 6.062326908111572, "genuineness": 4.143336772918701, "identity": 0.21787184476852417, "hypothesis": "Die Tür ist noch offen, ich habe deinen Mantel auf den Stuhl gelegt, sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/4900dbb136669906a267938ecde4f7eaa6145c642f76abf26090187c390470bc.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "492dfee25b8b78e9483e8391773965ae6dc74a0ec6b27e41c0f4b33371299144", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "f35c5b5d7ecb494daa568045b2c1315aa1ec842342168b0b7163ccf0a52d8a6d", "generation_seconds_batch": 20.22929605608806, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01454627513885498, "target_z": -0.562500021340835, "ce": 6.372091293334961, "genuineness": 4.036643981933594, "identity": 0.3241901397705078, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/492dfee25b8b78e9483e8391773965ae6dc74a0ec6b27e41c0f4b33371299144.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": "49312a21e7664b04aecbff1a6ab490e651242d957073bc9fc80f9558bbffba5d", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f33691c87869ae26dafaa31de6ad60e87afb68916ff8866d84362a5abf8a9941", "generation_seconds_batch": 17.156120534986258, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0059101879596710205, "target_z": -0.5234374724347548, "ce": 6.268836498260498, "genuineness": 4.168440818786621, "identity": 0.662266194820404, "hypothesis": "Ik schreef het nummer op de envelop, lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/49312a21e7664b04aecbff1a6ab490e651242d957073bc9fc80f9558bbffba5d.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "494a636f2ef22c5c7ff95a01c321ef73b7a5e523880effb49d86e22b02abc98e", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": 127, "seconds_generated": 10.16, "mp3_sha256": "3fe78f1f37bf35822df99a1ca23e8cfd150f131b510a16b8129b99dccd7378f9", "generation_seconds_batch": 21.659862173954025, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07443924248218536, "target_z": -0.39257811855328945, "ce": 6.268836498260498, "genuineness": 4.193545341491699, "identity": 0.5984541177749634, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/494a636f2ef22c5c7ff95a01c321ef73b7a5e523880effb49d86e22b02abc98e.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "495985a6c39999f1c8200d21299bc50edffe986f1ed9ec5696db06c1a81b1f67", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "b855e3fcb9358f83b984a2000e93985d7f7c66b7c7a926ae5bc40bf4112a51de", "generation_seconds_batch": 25.811485076905228, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00227239727973938, "target_z": -0.5390624919971869, "ce": 5.95112943649292, "genuineness": 4.042920112609863, "identity": 0.5209273099899292, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/495985a6c39999f1c8200d21299bc50edffe986f1ed9ec5696db06c1a81b1f67.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "49b9a96717e48c6524bc17bf925d59ed316bd6401c3a7d284a41f84cfe1ccfa2", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f8b3db60484b4c9ce03a58fd9d82bab61bbbec289a4ef9834fdf5b7774253381", "generation_seconds_batch": 26.641673012985848, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00227239727973938, "target_z": -0.5390624919971869, "ce": 6.229123115539551, "genuineness": 4.507347106933594, "identity": 0.5514317750930786, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/49b9a96717e48c6524bc17bf925d59ed316bd6401c3a7d284a41f84cfe1ccfa2.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "49d0394fb8c14fe4619cc254234b7a3622d07f2cd7a4b68669510e16288b34a3", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0b88882327d364596cb6bc8fcf1a14b365887fbfa2971676248493bec6c8620f", "generation_seconds_batch": 19.78877075400669, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.18694964051246643, "target_z": -0.17773438411431494, "ce": 6.221180438995361, "genuineness": 4.3127899169921875, "identity": 0.5938062071800232, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/49d0394fb8c14fe4619cc254234b7a3622d07f2cd7a4b68669510e16288b34a3.mp3" }, { "id": "heldout-es-0", "language": "es", "sentences": [ "Tu mensaje llegó después de que me fuera.", "Lo leí dos veces... y guardé el teléfono.", "Dime cuándo podemos hablar." ], "text": "Tu mensaje llegó después de que me fuera. Lo leí dos veces... y guardé el teléfono. Dime cuándo podemos hablar.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-0", "key": "49d468ae6540cc7c1e5b0e6883c25e2b29599458c9de0a5398000eaca62677ab", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Intense Malevolence Malice. 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": "102387e36df8b0a0ae140d254cc55f12739cb229c4d684d843803b5f1b75ae24", "generation_seconds_batch": 19.012101776083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.024774491786956787, "target_z": -0.5820312673394284, "ce": 6.205295085906982, "genuineness": 4.068024158477783, "identity": 0.46048814058303833, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/49d468ae6540cc7c1e5b0e6883c25e2b29599458c9de0a5398000eaca62677ab.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "4a399a834f075fbae25a7bbce7f08eacdc7a6a99194f144f5876cefb55a5304e", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "87915076f9378618d69bf5e1a312f04351b1f5c617e247daa364623564577d24", "generation_seconds_batch": 27.24132507899776, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07034794986248016, "target_z": -0.40039062833450545, "ce": 6.260893821716309, "genuineness": 3.9927117824554443, "identity": 0.5560855865478516, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/4a399a834f075fbae25a7bbce7f08eacdc7a6a99194f144f5876cefb55a5304e.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "4ac8eef284fca51e7c66f476d2bdad2332826df562e510a2a011135ff78066f7", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "01cb19efc70edd5f9c26e5b36946fc346c2594fad47c186876f00c0fa53d899a", "generation_seconds_batch": 25.70075239497237, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.17467579245567322, "target_z": -0.20117185654906974, "ce": 6.316492557525635, "genuineness": 3.9299514293670654, "identity": 0.6403155326843262, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/4ac8eef284fca51e7c66f476d2bdad2332826df562e510a2a011135ff78066f7.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "4ae78d8894866b5e02fc61df823dd9fee8f0e58401fc17bd6f63182e3ae40d7f", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "f5dca713fb7b35cf91fc7cfb761016e0b24b041c7773d9022e609cbaa8856628", "generation_seconds_batch": 25.36039623094257, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.035002708435058594, "target_z": -0.6015625133380219, "ce": 6.41180419921875, "genuineness": 4.2374773025512695, "identity": 0.3502766191959381, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/4ae78d8894866b5e02fc61df823dd9fee8f0e58401fc17bd6f63182e3ae40d7f.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": "4af01b66e5422b3f51f576a037c05a36ae8726410d4cf8ed931ced9d57c341fd", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "044ea2d2b852fcbd608322ef512af021a3b447f4dfadefc0d7435e01748dcde0", "generation_seconds_batch": 20.107833947986364, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.012500613927841187, "target_z": -0.5585937379957803, "ce": 6.173524379730225, "genuineness": 4.13078498840332, "identity": 0.5315179228782654, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue. Elle était ici, près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/4af01b66e5422b3f51f576a037c05a36ae8726410d4cf8ed931ced9d57c341fd.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "4b11e64ca968405883f553d074279c73885075aad921d374ba15d645c4a99369", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "4c68bfe1796cfd836d616f5c8d6c38363e1003dde24fe873d6fa23b4a37c5634", "generation_seconds_batch": 26.802801141981035, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0038645267486572266, "target_z": -0.5273437557798095, "ce": 6.252951145172119, "genuineness": 4.218649387359619, "identity": 0.15639567375183105, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/4b11e64ca968405883f553d074279c73885075aad921d374ba15d645c4a99369.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "4b2c93fede4b933cd08beb384e1616c5a22443b733711129474e7bd584bbe9d2", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "3480145c46036c64992e714662830255d349c34f6f0fcc13be9faa335efefd29", "generation_seconds_batch": 22.91845166904386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0038645267486572266, "target_z": -0.5273437557798095, "ce": 6.141753673553467, "genuineness": 4.582659721374512, "identity": 0.7884481549263, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/4b2c93fede4b933cd08beb384e1616c5a22443b733711129474e7bd584bbe9d2.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "4b888f44ebf363af8034f32a91335e3d1b210b61773a34d8f00234ed0b8fd1db", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "e337ff9419dcb5676942cc483c112068845f9267c82cacb7f3a869929707c901", "generation_seconds_batch": 27.29163707594853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.06364172697067261, "target_z": -0.6562500248976408, "ce": 6.403861999511719, "genuineness": 4.444586753845215, "identity": 0.37187257409095764, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/4b888f44ebf363af8034f32a91335e3d1b210b61773a34d8f00234ed0b8fd1db.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "4bb30d576cbf8ae566d67831bf74a574f4a499772512d32d53de2d9ef4f157ea", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1d8542757e67296e5feb6d475314a612f38a50fb1e22595175c8ab42acf8c078", "generation_seconds_batch": 27.325723235029727, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.012047111988067627, "target_z": -0.5117187362173774, "ce": 6.125868320465088, "genuineness": 4.256305694580078, "identity": 0.4787161350250244, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/4bb30d576cbf8ae566d67831bf74a574f4a499772512d32d53de2d9ef4f157ea.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "4bdc88c4d63e1e749603f3437efba6cbd678abf5fdf60d9fc577b272a10cc240", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "b0a23bd6f7977a504b81e4e4fc059d866dc348daf8b397a80f3baaba912f6457", "generation_seconds_batch": 18.795395730994642, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.016591906547546387, "target_z": -0.5664062477769963, "ce": 6.340320587158203, "genuineness": 4.293961524963379, "identity": 0.19557718932628632, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/4bdc88c4d63e1e749603f3437efba6cbd678abf5fdf60d9fc577b272a10cc240.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "4be122da93041f781a57d3a7dff21b967c1c0bfeb00f6673a3a00cb5b8958f6e", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "b434316fb87c576333a15ad0647f68f616c8d3c723ef2cb341ed3b65e03d80d7", "generation_seconds_batch": 25.91627170704305, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.023298144340515137, "target_z": -0.49023437700070327, "ce": 6.308549880981445, "genuineness": 4.042920112609863, "identity": 0.3208419382572174, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/4be122da93041f781a57d3a7dff21b967c1c0bfeb00f6673a3a00cb5b8958f6e.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "4c4336db0f780118dfb77bf2621c262b70fcdc2d9e261db30219dad24dc815ed", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "4487b0dc67007575af161860c5c523511d7f4badc3d1ec43a27204af61a254ea", "generation_seconds_batch": 25.811485076905228, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03659482300281525, "target_z": -0.46484376289342116, "ce": 5.863759994506836, "genuineness": 4.337893962860107, "identity": 0.8313769102096558, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/4c4336db0f780118dfb77bf2621c262b70fcdc2d9e261db30219dad24dc815ed.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "4c93c2d97b323855d20eb700876686e2a128e7bc83f6d8ab73eb2859293c2a65", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "caf7abd61cf6e4d14bb4e1330f9f322fb2b00ee4b738a4ade6ccbbc07283bc80", "generation_seconds_batch": 17.868277807021514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.014092743396759033, "target_z": -0.5078125097812161, "ce": 6.109982967376709, "genuineness": 4.369274139404297, "identity": 0.4076493978500366, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/4c93c2d97b323855d20eb700876686e2a128e7bc83f6d8ab73eb2859293c2a65.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "4cd3c9a42e9317863a725b4ac89049d8c2c0bdf5785b73ce8a286fba1df78478", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c77bdefe84d6ac242bf2204b32d7f3bdbfc62c83c1fcffda72c925d22059b267", "generation_seconds_batch": 25.418444887036458, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0778274536132812, "target_z": 1.5234374155258614, "ce": 6.213237762451172, "genuineness": 4.080576419830322, "identity": 0.17512273788452148, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/4cd3c9a42e9317863a725b4ac89049d8c2c0bdf5785b73ce8a286fba1df78478.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "4cff153109f50e92747f299f12928674aee930275e10f791b7dd1b152d455c56", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "10a8521aa1b418d2b6598fac377ac3ec81712037399c452107109b6b78360fd8", "generation_seconds_batch": 18.403548086993396, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.0452309250831604, "target_z": -0.6210937593366153, "ce": 6.268836498260498, "genuineness": 4.337893962860107, "identity": 0.452555388212204, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/4cff153109f50e92747f299f12928674aee930275e10f791b7dd1b152d455c56.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "4e3b2e235da86527804ecfa8e1a8c21e562b7b15c8403b4b4404802367d85893", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "e235320e44637553bcc0a22b3f64246471efb1579220b8dcdd8bc0a8a2b47636", "generation_seconds_batch": 27.049748205929063, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.05341348052024841, "target_z": -0.6367187219901541, "ce": 6.4515180587768555, "genuineness": 4.3881025314331055, "identity": 0.453594833612442, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/4e3b2e235da86527804ecfa8e1a8c21e562b7b15c8403b4b4404802367d85893.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "4e4ba01d2d9e42f75c5dbcd0a99084e0b042071498075a127b4c96a507365b09", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "d01724c6331df1a75c9ab2eb15d02c5c9d180e098e7784be8da05e71f3efee33", "generation_seconds_batch": 25.967064597061835, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.05136784911155701, "target_z": -0.6328124955539927, "ce": 6.348263263702393, "genuineness": 4.281409740447998, "identity": 0.3982139527797699, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/4e4ba01d2d9e42f75c5dbcd0a99084e0b042071498075a127b4c96a507365b09.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "4e6424e60dff9030c22089e181e912864993ed069564ea410b9919bf5ad996b5", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "bdf08a8b26a39ec78160f8318a08c9462d92b348f8b121aeec3cf969bb11b846", "generation_seconds_batch": 22.26370304694865, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.016138404607772827, "target_z": -0.5039062264361613, "ce": 6.372091293334961, "genuineness": 4.250029563903809, "identity": 0.4524354934692383, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/4e6424e60dff9030c22089e181e912864993ed069564ea410b9919bf5ad996b5.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "4ee687dd3d6f1e27f241b2311576e8d45f5cb67432d9604fcb69c02c8c1dfd36", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "460075dd8fc211fca6a8b0732996f758b3d205ed0f39aeede39960957184669b", "generation_seconds_batch": 18.668562631006353, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03250354528427124, "target_z": -0.47265624422019054, "ce": 5.657250881195068, "genuineness": 3.923675537109375, "identity": 0.2892725467681885, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/4ee687dd3d6f1e27f241b2311576e8d45f5cb67432d9604fcb69c02c8c1dfd36.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": "4f0da589963c70caca114fc69e0720aae851aa733f7a07d103144b54622aa693", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0d5ff0c7ec69c247a412ea8d22b321f44d63accecfc4f39dc1859156632417ec", "generation_seconds_batch": 19.05137082596775, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.018637537956237793, "target_z": -0.5703124742131577, "ce": 6.141753673553467, "genuineness": 4.362998008728027, "identity": 0.635528564453125, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí, junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/4f0da589963c70caca114fc69e0720aae851aa733f7a07d103144b54622aa693.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "4f0f6e8ffd741fe6746edc05a8961f1d07d455f44c99d386a82c2fc6c27319f4", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d101b4ba1ed39e9a844f9847875fb54f29f24517534e8dc96d9ae43451b4aa57", "generation_seconds_batch": 26.20452650100924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.016138404607772827, "target_z": -0.5039062264361613, "ce": 6.181467056274414, "genuineness": 4.1998209953308105, "identity": 0.7774652242660522, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/4f0f6e8ffd741fe6746edc05a8961f1d07d455f44c99d386a82c2fc6c27319f4.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "4f527e93c0d75b624692c3beb917b9356ee0fafcda7dadba78931b48b1603a72", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "3030fa05f6ba4164ec0403e9cfc0c9a5eb62fba5b343d552fa1f7bf261e580c9", "generation_seconds_batch": 24.031754920026287, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01045498251914978, "target_z": -0.5546875115596189, "ce": 6.3641486167907715, "genuineness": 4.005263805389404, "identity": 0.5178167819976807, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/4f527e93c0d75b624692c3beb917b9356ee0fafcda7dadba78931b48b1603a72.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": "4fc50bf85e4296b11abe87fb1b3caa4c7a7a940586b881f3ec5b81eb24e53b84", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "293e3af9fa11313319085096e9a55017d6b5f37186f251edb19da0ef52eb396d", "generation_seconds_batch": 17.217428064905107, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07750770449638367, "target_z": -0.38671875044460075, "ce": 6.165581703186035, "genuineness": 4.457139015197754, "identity": 0.7477846145629883, "hypothesis": "J'ai écrit le numéro sur l'enveloppe. Lis-le bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/4fc50bf85e4296b11abe87fb1b3caa4c7a7a940586b881f3ec5b81eb24e53b84.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": "4fccbd6fc92184b4d331d1cc6ce2d84f637e722a2aafaecdc255fa66db6c4a09", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7ae9d173bfceec5028b1092f42605717f0ddb0d373be8cf37166274209366c62", "generation_seconds_batch": 22.92426662403159, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02841225266456604, "target_z": -0.48046875400140654, "ce": 6.165581703186035, "genuineness": 4.124508857727051, "identity": 0.3716089129447937, "hypothesis": "Je bericht kwam nadat ik vertrokken was, ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/4fccbd6fc92184b4d331d1cc6ce2d84f637e722a2aafaecdc255fa66db6c4a09.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": "510ff63fd0e4a67de04d1eedf94082742fdb2478bb4cf418622080db2655447b", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5e66e3f0fa12a8a19540d6c6647aac1294e854f495f4ecf9011b353d7878157e", "generation_seconds_batch": 17.2846324989805, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.016138404607772827, "target_z": -0.5039062264361613, "ce": 6.324435234069824, "genuineness": 4.105680465698242, "identity": 0.550678551197052, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt, er stand hier neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/510ff63fd0e4a67de04d1eedf94082742fdb2478bb4cf418622080db2655447b.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": "51fc2a28750d0448b6df6c3fe34d247970251a0acb270c1df16b039fedb23226", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Escribí el número en el sobre.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "58aa8df1803db54d2152123eeac969f1d1f8cd33e7f5c6b53d09c73afe0cd6da", "generation_seconds_batch": 19.404513677000068, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.059096917510032654, "target_z": -0.4218749875511796, "ce": 6.165581703186035, "genuineness": 4.118232727050781, "identity": 0.6917270421981812, "hypothesis": "Escribí el número en el sobre, léelo bien, y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/51fc2a28750d0448b6df6c3fe34d247970251a0acb270c1df16b039fedb23226.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "520d06609dea870e99d36cbab9095603fea78ce5c6f0acb2c0013045c9694b97", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "dc9da84b98dfbd2953893481372c8f645c802dadfb6af64799dfd099a690e1d4", "generation_seconds_batch": 19.936476087081246, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.384674072265625, "target_z": 2.109375023119238, "ce": 6.038498878479004, "genuineness": 4.042920112609863, "identity": 0.5414435863494873, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/520d06609dea870e99d36cbab9095603fea78ce5c6f0acb2c0013045c9694b97.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "52ac81132196daf5b9973782689b132f6c7af51d31f25922d59c7fca749592eb", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "4eb15a202044a0826255d4facecb5e0cef671f37c70384b611553e49914ea940", "generation_seconds_batch": 24.96092503296677, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.29306742548942566, "target_z": 0.024902364257208622, "ce": 5.895530700683594, "genuineness": 4.011539936065674, "identity": 0.5931729078292847, "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/emo_Malevolence_Malice/52ac81132196daf5b9973782689b132f6c7af51d31f25922d59c7fca749592eb.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "530e2f010f9b17050f64c8f17d6ecc986c3e1d318fd0d654e35f39177614a590", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "d18dc118b6abb60acafd0d01deef8cab941ed641a4d7c84d12ae5fd129c6322e", "generation_seconds_batch": 27.181576865026727, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.0329570472240448, "target_z": -0.5976562299929672, "ce": 6.300607204437256, "genuineness": 4.3881025314331055, "identity": 0.4179125428199768, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/530e2f010f9b17050f64c8f17d6ecc986c3e1d318fd0d654e35f39177614a590.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": "531fb3dbc6c5d1b57110c5df51a96aad096bc443496cc3e9e0186a300d320b32", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6dcc63de0d394e6314f2dd9366be9001bdded73662a74dd6243773b0761603ce", "generation_seconds_batch": 24.735126861953177, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06114254891872406, "target_z": -0.41796876111501824, "ce": 6.24500846862793, "genuineness": 4.105680465698242, "identity": 0.7600882053375244, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar, wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/531fb3dbc6c5d1b57110c5df51a96aad096bc443496cc3e9e0186a300d320b32.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": "53696913b42a60d3758c21b7617c9c071238d23305bf767e529462efed6738f1", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "552dc2d01364e911b6bc2801fe5ed3a9c10268732c019ca064a30a84b2ffcf3e", "generation_seconds_batch": 22.437216124031693, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08466745913028717, "target_z": -0.373046872554696, "ce": 6.157639026641846, "genuineness": 4.105680465698242, "identity": 0.7787387371063232, "hypothesis": "J'ai écrit le numéro sur l'enveloppe, lis-le bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/53696913b42a60d3758c21b7617c9c071238d23305bf767e529462efed6738f1.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "536e527bf7c485739645040e2d297088f1a2a20b6752f01fe3a970bf2886910d", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "cd9cf35afa7d4242bec54e7c020073f8209647233a11e2e59698a1fd4c8f6544", "generation_seconds_batch": 24.82485379301943, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.055005624890327454, "target_z": -0.42968749733239564, "ce": 6.38003396987915, "genuineness": 3.986435890197754, "identity": 0.5496872663497925, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/536e527bf7c485739645040e2d297088f1a2a20b6752f01fe3a970bf2886910d.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "53b6d352ea118e70b47ec099f4372ab5f7bfa9805fd843544958369b14e87dae", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "e558e99a9c2ced349e313e288cd036c290278cd541788a6f5488997f67d7c201", "generation_seconds_batch": 28.6198508720845, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0038645267486572266, "target_z": -0.5273437557798095, "ce": 5.990842819213867, "genuineness": 4.042920112609863, "identity": 0.673164963722229, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/53b6d352ea118e70b47ec099f4372ab5f7bfa9805fd843544958369b14e87dae.mp3" }, { "id": "heldout-de-4", "language": "de", "sentences": [ "Ich schrieb die Nummer auf den Umschlag.", "Lies sie genau... und prüfe die letzten beiden Ziffern.", "Ruf an, wenn du Zeit hast." ], "text": "Ich schrieb die Nummer auf den Umschlag. Lies sie genau... und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-4", "key": "546203f143d2dbd908be3097fd76d41d76e93afd1487fa15c64a3d119dff5d49", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Intense Malevolence Malice. 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": "c762d8f8bdbcaf76387476c4692f05768b816ab50bb3ec986aa07d91831c49ec", "generation_seconds_batch": 19.06358708499465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.022275328636169434, "target_z": -0.49218749021878394, "ce": 6.308549880981445, "genuineness": 4.143336772918701, "identity": 0.6302112340927124, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, ließ sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/546203f143d2dbd908be3097fd76d41d76e93afd1487fa15c64a3d119dff5d49.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "551d9c24f80774171f78990664b681b158558bf9e54ef1a926da50ece9efbee9", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "9daf70cd43d9eac6285cfc802a4aa15df499cd84e2483ca6c810fabde6295544", "generation_seconds_batch": 29.797443560906686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6482423543930054, "target_z": 0.7031250835849371, "ce": 6.205295085906982, "genuineness": 4.256305694580078, "identity": 0.19128954410552979, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/551d9c24f80774171f78990664b681b158558bf9e54ef1a926da50ece9efbee9.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "5521da390fccb124b05e293abc4c441029e7d3ddfb1b130726f35e95b5c862d1", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "e2f4727238eba409e2967dfcc830a2dd61d9d65d84f9eabc1ae706f4e8cba1d3", "generation_seconds_batch": 28.34020373795647, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4053221344947815, "target_z": 0.2392578488461096, "ce": 6.268836498260498, "genuineness": 4.162164688110352, "identity": 0.5567564964294434, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/5521da390fccb124b05e293abc4c441029e7d3ddfb1b130726f35e95b5c862d1.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "556384b5fe6e409e639d80cb5d1a9f9018e35048bb7eaec1b62d6da4fdcdf3bb", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "c41bea7cbb646a7fd5367f5b39a03369deb44ac6b9d68c9a1aa169e65177885f", "generation_seconds_batch": 27.367997149005532, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00022676587104797363, "target_z": -0.5351562655610255, "ce": 6.165581703186035, "genuineness": 4.055472373962402, "identity": 0.704494833946228, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/556384b5fe6e409e639d80cb5d1a9f9018e35048bb7eaec1b62d6da4fdcdf3bb.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "5568b873f17211efee43d30799fb5edcf3d8ab421f4b87c7068e840fd9322dae", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "f3d2eec83687a9afb24bb96127f78876d1bbf2c0ee5802abe74f3d16f6a1dbb2", "generation_seconds_batch": 27.644933697069064, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0205495357513428, "target_z": 1.4140626200421968, "ce": 5.9431867599487305, "genuineness": 3.7416701316833496, "identity": 0.3621726930141449, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/5568b873f17211efee43d30799fb5edcf3d8ab421f4b87c7068e840fd9322dae.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "5577ce068a0b5923d280398cb381eaadac6dee1dfccbef62880ca7dcd1cfa71b", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9195aa20da19090d354bc03e994c902cc39097a7c9387646a5aaf7f87c11b5ac", "generation_seconds_batch": 33.55481519794557, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00022676587104797363, "target_z": -0.5351562655610255, "ce": 6.213237762451172, "genuineness": 4.582659721374512, "identity": 0.7176873683929443, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/5577ce068a0b5923d280398cb381eaadac6dee1dfccbef62880ca7dcd1cfa71b.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "558cdebec71f1c6352a387eb049fa79397097216db460f77a3c3410f1e2499a2", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "23953dc62e24f7fb679a7a60611b085cb06da59d2cce2a7154099c9f72897a28", "generation_seconds_batch": 25.27061033807695, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03454919159412384, "target_z": -0.4687499893295825, "ce": 6.173524379730225, "genuineness": 4.632867813110352, "identity": 0.6737876534461975, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/558cdebec71f1c6352a387eb049fa79397097216db460f77a3c3410f1e2499a2.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "55c731dad33efda9173ec649f7b44f8bf801f3844dd628228a4862ee0636bba8", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "5539a29c9a254a5f8352f2ddeef13a1006f7fedbfc1860f49138b13c4871e483", "generation_seconds_batch": 20.25811673491262, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.035002708435058594, "target_z": -0.6015625133380219, "ce": 6.38003396987915, "genuineness": 4.281409740447998, "identity": 0.2761325240135193, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/55c731dad33efda9173ec649f7b44f8bf801f3844dd628228a4862ee0636bba8.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "55cee46247ea6c7861558466d4666362291fda9a881e81c54b50b37be33a3d5d", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "80f3080e83fd1a8f494b90f23f7d1607c05db46648638e52bff9fe032c780e53", "generation_seconds_batch": 27.115666162921116, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.004318058490753174, "target_z": -0.5429687753422415, "ce": 5.935244083404541, "genuineness": 4.21237325668335, "identity": 0.7295869588851929, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/55cee46247ea6c7861558466d4666362291fda9a881e81c54b50b37be33a3d5d.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "55dd3a744ada5478557907352be37e37a9bcb130e398f6706516e278a795e25e", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "b2a20cff4f883c8e7d35330d7b97b3162c13be745ca20b223270ddaae3bc789f", "generation_seconds_batch": 25.179453340009786, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.004318058490753174, "target_z": -0.5429687753422415, "ce": 6.157639026641846, "genuineness": 4.369274139404297, "identity": 0.562827467918396, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/55dd3a744ada5478557907352be37e37a9bcb130e398f6706516e278a795e25e.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "55e63162813fe3b4248fca9d80c239ae7467f55a9645340d515ff3ff57ea16b1", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "dbe3f22dd625d4ec3b69283d7ab83d41e450847926591c33591c877de85ade7b", "generation_seconds_batch": 26.251836268929765, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06830230355262756, "target_z": -0.4042968832251135, "ce": 6.268836498260498, "genuineness": 4.20609712600708, "identity": 0.2282179594039917, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/55e63162813fe3b4248fca9d80c239ae7467f55a9645340d515ff3ff57ea16b1.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "55e67d7b148d23088a2df064d80be019930f2a1891e80f0a3e8ca0980a7e487f", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "f6a5a0394a4ef5e0076d51f0f90b42daaf463c88ff230a5d95525e91b17dda89", "generation_seconds_batch": 26.740647199912928, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03250354528427124, "target_z": -0.47265624422019054, "ce": 6.157639026641846, "genuineness": 4.519899368286133, "identity": 0.3661917448043823, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/55e67d7b148d23088a2df064d80be019930f2a1891e80f0a3e8ca0980a7e487f.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "55f49073e0e89746da3ebe0ed8b708d5517fdcd1ae39603a48dd125c9c256f73", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "cd83a83aa3f1384fcece6db1ef67eab03611c87d986824f19044b71e8e6da682", "generation_seconds_batch": 25.513310719979927, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.007955819368362427, "target_z": -0.5195312459985935, "ce": 6.165581703186035, "genuineness": 4.15588903427124, "identity": 0.5903513431549072, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/55f49073e0e89746da3ebe0ed8b708d5517fdcd1ae39603a48dd125c9c256f73.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "5605d2913fccfa2ad6d1a913142bff0d049c7171d117a073dbe3f6a48a2d1fd7", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "050dadcfa8265991b90d127f595a70bd93b7d998c6b0ac141b156d346413708e", "generation_seconds_batch": 20.780900493031368, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00636368989944458, "target_z": -0.5468750017784029, "ce": 6.109982967376709, "genuineness": 3.8420867919921875, "identity": 0.3873211145401001, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/5605d2913fccfa2ad6d1a913142bff0d049c7171d117a073dbe3f6a48a2d1fd7.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "563fed26373b04a0d362c257f159965e70d7a2eacde52ec8ebf1040d6bad891c", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "9ef0cb19dff2fcceb176621a558c73e340fa8112f85507b2ff28f343bdc424ee", "generation_seconds_batch": 18.69529991107993, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.014092743396759033, "target_z": -0.5078125097812161, "ce": 6.24500846862793, "genuineness": 4.168440818786621, "identity": 0.3057863116264343, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/563fed26373b04a0d362c257f159965e70d7a2eacde52ec8ebf1040d6bad891c.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "56c03519a31631e06eb0c7f3b8b1da5eb33f67cd30349a76575e06a06810ff2c", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "d6e19802edc9bf66e2e3efc55348864964a88a1bdde1874a5537ad49ae5262d3", "generation_seconds_batch": 27.049748205929063, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.04727655649185181, "target_z": -0.6249999857727767, "ce": 6.300607204437256, "genuineness": 4.243753433227539, "identity": 0.44500070810317993, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/56c03519a31631e06eb0c7f3b8b1da5eb33f67cd30349a76575e06a06810ff2c.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "56f3a0ebf36c79aa80f5e3e7965b2bb61242caf97e05fc5097602495efde89d5", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "391e2226fba28de189b25117330f5bdec8086e518f557186938ecedd5d9edb84", "generation_seconds_batch": 20.6772598689422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07137078046798706, "target_z": -0.39843748666197815, "ce": 5.736677646636963, "genuineness": 3.9425034523010254, "identity": 0.5065385103225708, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/56f3a0ebf36c79aa80f5e3e7965b2bb61242caf97e05fc5097602495efde89d5.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "572ef96d2ea7647478d8b862647195b0a81232ca4b762a30f0fd26684ca078e6", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "3bb711c36746a7a0598d17abacf512649fffdbb4265a3076dcb8e960e4594e8a", "generation_seconds_batch": 28.61614420299884, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03557200729846954, "target_z": -0.46679687611150184, "ce": 5.863759994506836, "genuineness": 4.2751336097717285, "identity": 0.7577342987060547, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/572ef96d2ea7647478d8b862647195b0a81232ca4b762a30f0fd26684ca078e6.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "574a5d425acd96bc4f2d0161ec9cdb62d17df49b4f9a3b04759a660570b40b90", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "1e5d5881cca3dd007aa4eb929ee91ab2379162be7965a95d1d92e2f93fcd3bba", "generation_seconds_batch": 25.749805856030434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1480824202299118, "target_z": -0.25195311321808067, "ce": 6.24500846862793, "genuineness": 3.9173994064331055, "identity": 0.114749014377594, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/574a5d425acd96bc4f2d0161ec9cdb62d17df49b4f9a3b04759a660570b40b90.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": "57b867d32b90fc95e949d76637301b5fc5f96bbcca7abba9845912c12719b770", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b9f15afd9e56d7da430828fd01492c1aa9f5e6c6e730fce08e9644209dad61fa", "generation_seconds_batch": 19.460877311998047, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.029435068368911743, "target_z": -0.47851564078332587, "ce": 6.24500846862793, "genuineness": 4.080576419830322, "identity": 0.6566267013549805, "hypothesis": "Ik schreef het nummer op de envelop. Lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/57b867d32b90fc95e949d76637301b5fc5f96bbcca7abba9845912c12719b770.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "57e12c97e61b7b493ef84e2c710551aca65b4cb0906a68d9caf087f46381bf0c", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9017e63244bb96e6f9049c9b0c2979af8eaf55c2e68f631562dfbb425dcaad94", "generation_seconds_batch": 19.298494191025384, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.29907649755477905, "target_z": 0.036376961322325936, "ce": 5.557967185974121, "genuineness": 4.093128204345703, "identity": 0.44741541147232056, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/57e12c97e61b7b493ef84e2c710551aca65b4cb0906a68d9caf087f46381bf0c.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": "588e2b9b6b88d5bf0d75d1c137d78158f2b7dbddc94a9b81a5bfbe996d8a1afd", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Someone moved the little blue chair.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8eca6099d7985fcbb4e43cb8da9b4ef27b96573ae043b3d98337ccadd876a266", "generation_seconds_batch": 17.13475025992375, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.012500613927841187, "target_z": -0.5585937379957803, "ce": 6.38797664642334, "genuineness": 4.306513786315918, "identity": 0.38482666015625, "hypothesis": "Someone moved the little blue chair. It used to stand here beside the tall bookshelf. We can put it back after lunch.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/588e2b9b6b88d5bf0d75d1c137d78158f2b7dbddc94a9b81a5bfbe996d8a1afd.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "5895d5644ee14299394b5bf0e4e86925e0e3f79284f585325dc8a1c6afc705a8", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "0dbb6873864dd7037c506d70f8d3905ef78480a5866e66bcc7564762e913700e", "generation_seconds_batch": 25.255807731067762, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.08409816026687622, "target_z": -0.6953125168948276, "ce": 6.3641486167907715, "genuineness": 4.218649387359619, "identity": 0.4381901025772095, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/5895d5644ee14299394b5bf0e4e86925e0e3f79284f585325dc8a1c6afc705a8.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "58eab43a9d60e266f7ab852b6ef5016f9c02e139ca18aab5cdaea938b355839e", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "573dd0d9ebedcc9f62edf6facf37c853499e5b8c632396169faa8e82d10a25ba", "generation_seconds_batch": 26.848008858039975, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1368313729763031, "target_z": -0.27343750088920143, "ce": 6.157639026641846, "genuineness": 4.061748504638672, "identity": 0.5053699016571045, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/58eab43a9d60e266f7ab852b6ef5016f9c02e139ca18aab5cdaea938b355839e.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "5930ad66366eae1d2daacc5699ef8273292133deb1ff98d6ebcbf86fccdc2a6e", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "b5fa62a1145d5650669db512fa9b354c21dde7f6874862e3cd5e1ea3df700f50", "generation_seconds_batch": 24.20346673997119, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0038645267486572266, "target_z": -0.5273437557798095, "ce": 6.300607204437256, "genuineness": 3.973883628845215, "identity": 0.4314384460449219, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/5930ad66366eae1d2daacc5699ef8273292133deb1ff98d6ebcbf86fccdc2a6e.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "595e632f771fe4102ecec966e5aa5ff629f6c87bdf19a572e52ac0622a002c1b", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "1143f51ffeb8718211073acbcbe7643eec90476fef60f8f2492e4f9af5101f03", "generation_seconds_batch": 25.8792836680077, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.03704833984375, "target_z": -0.6054687397741833, "ce": 6.1894097328186035, "genuineness": 4.344170093536377, "identity": 0.7473568320274353, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/595e632f771fe4102ecec966e5aa5ff629f6c87bdf19a572e52ac0622a002c1b.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "5aa48c548aaaec5c09c1468c2e9ed7ddf10a217afefb384f968aacf772c2fd12", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f9e364342698458458ca057f1ad883fce11551b0f607d0c5be6349f7fc5de0ef", "generation_seconds_batch": 26.067440679995343, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03557200729846954, "target_z": -0.46679687611150184, "ce": 5.748591423034668, "genuineness": 3.8546390533447266, "identity": 0.3582274913787842, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/5aa48c548aaaec5c09c1468c2e9ed7ddf10a217afefb384f968aacf772c2fd12.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "5ab8ad40f9a288b482dd05ab3bcdb9984bd0409190a48d18b19f4d9acf3e1fd9", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "11fee44b3282699f1659010bc93f22cf41b1fc91fd805aed0c7ea42a3b613224", "generation_seconds_batch": 25.97593635006342, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06523384153842926, "target_z": -0.4101562513338022, "ce": 6.332377910614014, "genuineness": 4.193545341491699, "identity": 0.49366462230682373, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/5ab8ad40f9a288b482dd05ab3bcdb9984bd0409190a48d18b19f4d9acf3e1fd9.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "5af2feb0903b4140292e5dc083d32ccf56565766706253679fa60ba1bd4807ef", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "9f8695c5b00c8429034820c71212329e36d1c920e0a773aed6afb99406001338", "generation_seconds_batch": 24.957745958003215, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.0329570472240448, "target_z": -0.5976562299929672, "ce": 6.213237762451172, "genuineness": 4.457139015197754, "identity": 0.6463637351989746, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/5af2feb0903b4140292e5dc083d32ccf56565766706253679fa60ba1bd4807ef.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "5b0ee222fa68113ca681ba859f826faf87851a6b494fa74f03eab622a00dde0f", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "00b2015bf9b152e8fd75253a82d2d42e80349fe85c17e18aee53ce0147029262", "generation_seconds_batch": 22.9304855678929, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.028865784406661987, "target_z": -0.5898437771206445, "ce": 6.38003396987915, "genuineness": 4.149612903594971, "identity": 0.3706240653991699, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/5b0ee222fa68113ca681ba859f826faf87851a6b494fa74f03eab622a00dde0f.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "5b366485c990fa6aa69dd87d64c212fce23e346e03e8c5ad09144328fd509894", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "805e42997878230a2f2875bb56357179cf54c7cb6475e1146233a91309a6fbc8", "generation_seconds_batch": 25.97454882506281, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.039094001054763794, "target_z": -0.6093750231192379, "ce": 6.260893821716309, "genuineness": 4.18099308013916, "identity": 0.47107404470443726, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/5b366485c990fa6aa69dd87d64c212fce23e346e03e8c5ad09144328fd509894.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "5b8710feb159b8ee05f4008b8a28f2e37d5302e189987a17426209cbf92a74b8", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "fd6039551dd1dff91a2b0ac0077468f063542b1cffd117d93804137ed756f4d2", "generation_seconds_batch": 25.957190423971042, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.007955819368362427, "target_z": -0.5195312459985935, "ce": 6.173524379730225, "genuineness": 4.331617832183838, "identity": 0.7323527336120605, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/5b8710feb159b8ee05f4008b8a28f2e37d5302e189987a17426209cbf92a74b8.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "5bfce46d60b0060bb238be8d9756f28c0834a29cf594811c02a39b9a9186d1fd", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "bd044a502b7d61b5c005fbe5432851763dad116311b07cb0af96e491705944f6", "generation_seconds_batch": 24.60730907099787, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.039094001054763794, "target_z": -0.6093750231192379, "ce": 6.197352409362793, "genuineness": 4.6203155517578125, "identity": 0.5252597332000732, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/5bfce46d60b0060bb238be8d9756f28c0834a29cf594811c02a39b9a9186d1fd.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": "5c7af30af4eff81c05381710a79dc86d17b74bbe57aea156b2e71476e4de99c2", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Your message arrived after I had left.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ef0c7cecb4def3dfd3a76ae78621f30df6b32d6ca300b2d0b1ad78c6af35a2f6", "generation_seconds_batch": 19.421222764067352, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0059101879596710205, "target_z": -0.5234374724347548, "ce": 6.467403411865234, "genuineness": 3.9927117824554443, "identity": 0.15843726694583893, "hypothesis": "Your message arrived after I had left. I read it twice, before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/5c7af30af4eff81c05381710a79dc86d17b74bbe57aea156b2e71476e4de99c2.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "5d74ee9f7bb602239c4457581ddd02b0140de01e85e90f75a90589809f49e57b", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "6159972b06b9e02bc44affbd4a435f8458107ecb6521b7c754fcb0f091f04a31", "generation_seconds_batch": 26.97431276203133, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00022676587104797363, "target_z": -0.5351562655610255, "ce": 6.09409761428833, "genuineness": 4.055472373962402, "identity": 0.6001909971237183, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/5d74ee9f7bb602239c4457581ddd02b0140de01e85e90f75a90589809f49e57b.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": "5e0458db4def2830ed4b8c823c299eeacbf07de909664f04c9763b7de4b2a1ce", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Your message arrived after I had left.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2adad7bc96fdcb7344f5d6627a68e69d7ccdb8ae895d28e99b9601a1e526ca56", "generation_seconds_batch": 27.025717504904605, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05091433227062225, "target_z": -0.43750000711361164, "ce": 6.427689552307129, "genuineness": 4.105680465698242, "identity": 0.4061860740184784, "hypothesis": "Your message arrived after I had left. I read it twice before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/5e0458db4def2830ed4b8c823c299eeacbf07de909664f04c9763b7de4b2a1ce.mp3" }, { "id": "heldout-fr-2", "language": "fr", "sentences": [ "La route tourne à gauche après le pont.", "Continue... jusqu’à la petite barrière en bois.", "Je t’attendrai de l’autre côté." ], "text": "La route tourne à gauche après le pont. Continue... jusqu’à la petite barrière en bois. Je t’attendrai de l’autre côté.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-2", "key": "5e0af124ad7421917e897f2876d49b25e213a1bc87e01fe43f09b9c255c2a32f", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Intense Malevolence Malice. 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": "edfafe76e373448931672b7fbc866c46b37817336cbee4a1f1b15d6c6061d13f", "generation_seconds_batch": 18.662544685998, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04886868596076965, "target_z": -0.4414062620042197, "ce": 6.2767791748046875, "genuineness": 4.444586753845215, "identity": 0.837344765663147, "hypothesis": "La route tourne à gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/5e0af124ad7421917e897f2876d49b25e213a1bc87e01fe43f09b9c255c2a32f.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "5e349078fc678e9116a65af3db217d857fb1a726a830d07a83307c5264d0efbc", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "e77b5db23f8187c4b5539e0a115d52d352c77c9c8193b1dbb293f7c1f3230f74", "generation_seconds_batch": 28.163135805050842, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.26366129517555237, "target_z": -0.031249982215970844, "ce": 5.911416053771973, "genuineness": 4.080576419830322, "identity": 0.6118915677070618, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/5e349078fc678e9116a65af3db217d857fb1a726a830d07a83307c5264d0efbc.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "5e597d68f5564110e03c5002dfb7a0b9ed7a4f3058eb4b28a44f4abf6ebe357e", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "9bc40c97e75e05a349744cdb453f1f252f5d5552453596af3d7baa4dfa63f6d5", "generation_seconds_batch": 26.32882586598862, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04068611562252045, "target_z": -0.4570312531122051, "ce": 6.229123115539551, "genuineness": 4.306513786315918, "identity": 0.30385786294937134, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/5e597d68f5564110e03c5002dfb7a0b9ed7a4f3058eb4b28a44f4abf6ebe357e.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": "5e5f3119960da6d3480f1c7750fc08d457f08b5463b5507576dac538e5a20c8e", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2c528662f1702abe1aa4de51f2ad2e0342eb91f478e6f22d54b30a0b232735e7", "generation_seconds_batch": 23.071617268957198, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0059101879596710205, "target_z": -0.5234374724347548, "ce": 6.141753673553467, "genuineness": 4.093128204345703, "identity": 0.7377406358718872, "hypothesis": "Ik schreef het nummer op de envelop, lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/5e5f3119960da6d3480f1c7750fc08d457f08b5463b5507576dac538e5a20c8e.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "5e8320a663622f73896ba432a84ab4e6a25e181e544421aa7d89ec887e8b14c7", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "b54f85ae32d27e65f4306fc239dc3429f8028f39d987f918fcd568b925d24aa1", "generation_seconds_batch": 26.666601025965065, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.016138404607772827, "target_z": -0.5039062264361613, "ce": 6.197352409362793, "genuineness": 4.281409740447998, "identity": 0.7665796279907227, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/5e8320a663622f73896ba432a84ab4e6a25e181e544421aa7d89ec887e8b14c7.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "5eb03aeb2752f07470adf0f835a689ce7eced9776088a2851cf7f54bc8dd4b83", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "c6de715b96f4008ef9cb9cdc21bbc9b90b7f44faf518210499e48aa1220ad1eb", "generation_seconds_batch": 27.838630085927434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5091385245323181, "target_z": 0.437499978659165, "ce": 6.1894097328186035, "genuineness": 4.293961524963379, "identity": 0.36742472648620605, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/5eb03aeb2752f07470adf0f835a689ce7eced9776088a2851cf7f54bc8dd4b83.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "5fe14cb3d56d1a5eec6b9f95fc559afce10fb803c92d5f4ff7e18cd8056754f3", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "f5548bc4ce100583da86673136af6bc40f7176db2d5ed576a7b17850c7a5406b", "generation_seconds_batch": 26.32882586598862, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.012500613927841187, "target_z": -0.5585937379957803, "ce": 6.125868320465088, "genuineness": 4.218649387359619, "identity": 0.34642258286476135, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/5fe14cb3d56d1a5eec6b9f95fc559afce10fb803c92d5f4ff7e18cd8056754f3.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "5ff0ab83c35463ea0a98039172f814933219d51a757e2d51277ed2985cde79ed", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "9e046e5fcb6edbefcc9d7488b04c339dd951fb67506d41626bb22444a77cee55", "generation_seconds_batch": 28.34020373795647, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03045789897441864, "target_z": -0.47656249911079857, "ce": 6.300607204437256, "genuineness": 4.381826400756836, "identity": 0.3940231204032898, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/5ff0ab83c35463ea0a98039172f814933219d51a757e2d51277ed2985cde79ed.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": "6011776e6eb13c7b70e52495a2dc3d27ad9ba9d929ee0f65c808c0df90b89c75", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4f3879c5c9cb4d4c7cf9cf0629ae6d3f383409e83c4b81d6bd8d368eb97eaa64", "generation_seconds_batch": 17.217428064905107, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09591849148273468, "target_z": -0.35156251333802185, "ce": 5.967014789581299, "genuineness": 4.306513786315918, "identity": 0.6148922443389893, "hypothesis": "J'ai écrit le numéro sur l'enveloppe. Lis-le bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/6011776e6eb13c7b70e52495a2dc3d27ad9ba9d929ee0f65c808c0df90b89c75.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "6027f92d280932608ee9ecd99fbf537531761132124ff8388e9ef374ff9dd668", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "27a3b9c2e17b306349e641990ee2bddf1f837da06c564754c290c5bcd489b311", "generation_seconds_batch": 19.406625418108888, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04886868596076965, "target_z": -0.4414062620042197, "ce": 6.181467056274414, "genuineness": 3.9299514293670654, "identity": 0.24114742875099182, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/6027f92d280932608ee9ecd99fbf537531761132124ff8388e9ef374ff9dd668.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "60558c54abebd4dc7a452dcc8db2aa5e48453ca42b8e99096c3226e6d152b1dc", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0f7bf9fe888c7e4ad50a75c1d6f85d060bd9c32437ed2056f08dbda69109864c", "generation_seconds_batch": 27.873861138010398, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10716953873634338, "target_z": -0.3300781256669011, "ce": 6.109982967376709, "genuineness": 4.18099308013916, "identity": 0.7004899978637695, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/60558c54abebd4dc7a452dcc8db2aa5e48453ca42b8e99096c3226e6d152b1dc.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "605bd81f1b399a56bc372d69bbb5987ab757c8fc4f9321334cded496ea905347", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "c24596cb0548ef501d7e71caaf143d51eec16b952458d7abaf6ca5007a3acb3b", "generation_seconds_batch": 29.620181945036165, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3485555052757263, "target_z": 0.13085938233591202, "ce": 6.0146708488464355, "genuineness": 3.8922953605651855, "identity": 0.24832676351070404, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/605bd81f1b399a56bc372d69bbb5987ab757c8fc4f9321334cded496ea905347.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "6064abfc5f78358331e1e6da210968cc1e14e27909ee61886fa80ceb6c1202c7", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "4fa766dd19e1ca16e59ae09102ca873aece30bbfea0bd45772a35330aa5d4526", "generation_seconds_batch": 27.180471549043432, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3654320538043976, "target_z": 0.16308592116092321, "ce": 6.213237762451172, "genuineness": 4.406930446624756, "identity": 0.6109845638275146, "hypothesis": "Die Tür ist noch offen, ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/6064abfc5f78358331e1e6da210968cc1e14e27909ee61886fa80ceb6c1202c7.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "6067c6fe9eceb414db42fc1755717f23e80c2709d378f57d8d4a0a749da5090e", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "1b06241b9a38c57d5b01b1f1987801920858c57e61ee173cf11019c3466f9fba", "generation_seconds_batch": 25.1878926520003, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01454627513885498, "target_z": -0.562500021340835, "ce": 6.221180438995361, "genuineness": 4.281409740447998, "identity": 0.6968591213226318, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/6067c6fe9eceb414db42fc1755717f23e80c2709d378f57d8d4a0a749da5090e.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "60b185c63b8163ba89502325c8e572bc4e9e915725f1f0545a4e162ce969a71e", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "e39808e0ea7465bc4b5b46496e7180d017e36886b05078bc76f3f395b9b3646c", "generation_seconds_batch": 25.002688780077733, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.33270177245140076, "target_z": 0.10058595672898153, "ce": 5.895530700683594, "genuineness": 4.093128204345703, "identity": 0.36409685015678406, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/60b185c63b8163ba89502325c8e572bc4e9e915725f1f0545a4e162ce969a71e.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "61378de7a2a28b871f1b1dac804f07ea656751c6a506e0cf96cc2dccd05aa09c", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "80cad390d7b05467cac98e38af32e25294522b8047de282a1d926dd08bf00eb9", "generation_seconds_batch": 27.181576865026727, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.05955043435096741, "target_z": -0.6484375151164248, "ce": 6.308549880981445, "genuineness": 4.20609712600708, "identity": 0.36019325256347656, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/61378de7a2a28b871f1b1dac804f07ea656751c6a506e0cf96cc2dccd05aa09c.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "61393e80dd549cee10ec610f0e77902f25bb197da9a0e2577a30590c93a8562e", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "6e1b3a6c7e9cb8bb6111995f98039819795843b3fca87053512ef49169f54709", "generation_seconds_batch": 21.216272560996003, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00636368989944458, "target_z": -0.5468750017784029, "ce": 6.133810997009277, "genuineness": 4.494794845581055, "identity": 0.7231583595275879, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/61393e80dd549cee10ec610f0e77902f25bb197da9a0e2577a30590c93a8562e.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "618c60802bbd28dd26d34824f280ca2d779d7865657cccfc97b2318662c1a2a7", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "ea155d9cefcfc5af45f1f4ea9acf8f670ac4f09a8691c928b8d398e88a952e77", "generation_seconds_batch": 26.388338941033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06421102583408356, "target_z": -0.41210936455188285, "ce": 6.3641486167907715, "genuineness": 4.243753433227539, "identity": 0.5055786371231079, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/618c60802bbd28dd26d34824f280ca2d779d7865657cccfc97b2318662c1a2a7.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "61bf43bb1c7921bc21af28ccd98874f35240221c2982678929016e7bbec9a1c0", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "05bcf06009d4c8680ee512f4f0670cb2a41ad122bd1e03bee59d496c18f9fb8d", "generation_seconds_batch": 26.34775253199041, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02432096004486084, "target_z": -0.4882812637826226, "ce": 6.0305562019348145, "genuineness": 4.055472373962402, "identity": 0.5165746212005615, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/61bf43bb1c7921bc21af28ccd98874f35240221c2982678929016e7bbec9a1c0.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "621b1361f20171ef2dc04693c40e2d86d6a8b3feced069f1cc9fbab96fcae638", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "1182e4ce450f211b08171d61021efa59dc5660829198e6b3dc539773e18faa71", "generation_seconds_batch": 29.301259291009046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09182721376419067, "target_z": -0.3593749946647913, "ce": 5.911416053771973, "genuineness": 4.168440818786621, "identity": 0.3431738615036011, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/621b1361f20171ef2dc04693c40e2d86d6a8b3feced069f1cc9fbab96fcae638.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "622169bebb7e7d2c7afb74f28fe1f8ebc53f7858db40c36f7727ee81220dceee", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "52a11f1a37549b01f723326c7df10f12cc307f18f959be57dbeeab93033766f0", "generation_seconds_batch": 27.577267786022276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.007955819368362427, "target_z": -0.5195312459985935, "ce": 6.165581703186035, "genuineness": 3.8295347690582275, "identity": 0.13173630833625793, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/622169bebb7e7d2c7afb74f28fe1f8ebc53f7858db40c36f7727ee81220dceee.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": "6251021e965909da18c372e0082deb48ac5201318df32281e168caafe4693254", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "ef269f23707ec1d2b4ba17cbc387662d0ba1c0c462a38cdcfbf16648370a975e", "generation_seconds_batch": 17.443555210949853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00022676587104797363, "target_z": -0.5351562655610255, "ce": 6.4515180587768555, "genuineness": 4.306513786315918, "identity": 0.5082359313964844, "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/emo_Malevolence_Malice/6251021e965909da18c372e0082deb48ac5201318df32281e168caafe4693254.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "6277e9e64577f05b8894c6fb7b90508e06aed2d2cd09826a532b2f4257f45722", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "c4c21cb5c372da6d85399b9b505e27b080bfea3e125249b4122d5acbb5b6a245", "generation_seconds_batch": 24.424292281037197, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.34497562050819397, "target_z": 0.12402342916373633, "ce": 5.7247633934021, "genuineness": 3.8671910762786865, "identity": 0.4857313632965088, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/6277e9e64577f05b8894c6fb7b90508e06aed2d2cd09826a532b2f4257f45722.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "627a68c69acbb28b9a460195f08a143ef226aa33de683c133c21dca9ee1b3e20", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b912052d2b8e71cb7e96b4e012ee451320ddaab006a1dec7944a546851663373", "generation_seconds_batch": 24.177027603029273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0328233242034912, "target_z": 1.4374999786591651, "ce": 6.348263263702393, "genuineness": 4.168440818786621, "identity": 0.009152932092547417, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/627a68c69acbb28b9a460195f08a143ef226aa33de683c133c21dca9ee1b3e20.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "62e25d5216deeae26a391f9cc8042a79b8001eae2a97a2d06c19c40c30315aa7", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "0c85495d7745b511550721335d85f7ecf2affb488a101122e44a3dfbd887b1b9", "generation_seconds_batch": 27.29163707594853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.05750477313995361, "target_z": -0.6445312317713701, "ce": 6.300607204437256, "genuineness": 4.256305694580078, "identity": 0.4747067093849182, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/62e25d5216deeae26a391f9cc8042a79b8001eae2a97a2d06c19c40c30315aa7.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": "6317f96ce3fcb79c53ab57e95a8b85da2fc255e3eddf776ae9cb94248839a34a", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a6c81a33aab88c0fe39b38ac7983a768a05142c6010d5fd06877625a955b8ec0", "generation_seconds_batch": 25.603216346004046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09489567577838898, "target_z": -0.3535156265561025, "ce": 6.308549880981445, "genuineness": 4.344170093536377, "identity": 0.4344605803489685, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/6317f96ce3fcb79c53ab57e95a8b85da2fc255e3eddf776ae9cb94248839a34a.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": "6321ac8d4e0487a07836f1c6032ab04342c3a518ad0f40d0d90d25497df91be4", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8475b100c9d8cce1af753a2259026aa9965cf64d88750b1b7abd83b865ac4189", "generation_seconds_batch": 17.239664223976433, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.014092743396759033, "target_z": -0.5078125097812161, "ce": 6.316492557525635, "genuineness": 4.218649387359619, "identity": 0.49670881032943726, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/6321ac8d4e0487a07836f1c6032ab04342c3a518ad0f40d0d90d25497df91be4.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "6380d28c51dd6e9c8d7bc4fa892a31ae5fd6aeaddd315fbda199ee96f9ab0bf3", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "383babb154177865c892f26da849f7cbbc3f1724a845c9a9ab0b28a76774aeb5", "generation_seconds_batch": 20.740487104980275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03864046931266785, "target_z": -0.46093750800281313, "ce": 6.070269584655762, "genuineness": 4.457139015197754, "identity": 0.7857974767684937, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/6380d28c51dd6e9c8d7bc4fa892a31ae5fd6aeaddd315fbda199ee96f9ab0bf3.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "63efffe39ce1c168cf0923d717e061e1121918701d6cab7999b3fbc664470fae", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "54bf0a81daa952dadb053c6caef06e453ef6d27884201a6d5acba3e2912f6ee0", "generation_seconds_batch": 25.04809733503498, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06932513415813446, "target_z": -0.4023437415525862, "ce": 5.80021858215332, "genuineness": 4.168440818786621, "identity": 0.6523653268814087, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/63efffe39ce1c168cf0923d717e061e1121918701d6cab7999b3fbc664470fae.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "645dca6a292ca38f105c19e2adee685e42bac3b615d377bf43a312aefe3e29ab", "bucket": "emo_Malevolence_Malice", "phase": "wording", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "f368afa5dad7c9cdd4a849a02744d7207de003a5da243dc99caf34c04449d408", "generation_seconds_batch": 19.26459373498801, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01045498251914978, "target_z": -0.5546875115596189, "ce": 6.038498878479004, "genuineness": 4.231201171875, "identity": 0.640445351600647, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/645dca6a292ca38f105c19e2adee685e42bac3b615d377bf43a312aefe3e29ab.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "6498959e211d808b8858c8c1968342048b9d253ef903480b88e1f179602f58a2", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "019f3e77c098aaed81001dc04f9a98d983678ce129738a6ef90b359411a73483", "generation_seconds_batch": 25.15201999794226, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02432096004486084, "target_z": -0.4882812637826226, "ce": 6.109982967376709, "genuineness": 4.011539936065674, "identity": 0.2662753164768219, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/6498959e211d808b8858c8c1968342048b9d253ef903480b88e1f179602f58a2.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "650113c13364af0909e7b5e22d09cc239ed0c06afcc1a988af0f8517b997f29c", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3e77d6c9c70360c38ec29576ff59dbdf25f69c7040bc591a57fe9ca416dfc935", "generation_seconds_batch": 27.190540881012566, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11637493968009949, "target_z": -0.31249999288638836, "ce": 6.125868320465088, "genuineness": 4.331617832183838, "identity": 0.5657775402069092, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/650113c13364af0909e7b5e22d09cc239ed0c06afcc1a988af0f8517b997f29c.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "652d4908ed02675208269f4da347aef868b89f2eedb37cf370102e94adb6046e", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "6e4cacfd044d1c5aa93d0c9b765f1a433f1d782c2a664add17769e8aced1044b", "generation_seconds_batch": 18.668562631006353, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1637444496154785, "target_z": 1.6874998363869318, "ce": 5.7684478759765625, "genuineness": 3.886019229888916, "identity": 0.4230647683143616, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/652d4908ed02675208269f4da347aef868b89f2eedb37cf370102e94adb6046e.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "652d7b50dcd3455bafd8cb2fae1cfdb248e90ab63fc873e3572c32607c49d8d1", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "374eb1c31ddadc0243a3c76ed8b1719135635bf1b172c700c63d3716d2919938", "generation_seconds_batch": 26.56255738704931, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09489567577838898, "target_z": -0.3535156265561025, "ce": 5.391171455383301, "genuineness": 3.8671910762786865, "identity": 0.4453534483909607, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/652d7b50dcd3455bafd8cb2fae1cfdb248e90ab63fc873e3572c32607c49d8d1.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "65431e7a341ad8029f403cd35f96b564799fd2d5286afd22c8ff1d72c51de0a5", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "767cfe952369c39ed27b6f8fe7452fab1744b85c409d7f3ddce8eedc28d08da1", "generation_seconds_batch": 23.933765553985722, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00636368989944458, "target_z": -0.5468750017784029, "ce": 6.213237762451172, "genuineness": 4.118232727050781, "identity": 0.6570985317230225, "hypothesis": "J'ai trouvé en en la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/65431e7a341ad8029f403cd35f96b564799fd2d5286afd22c8ff1d72c51de0a5.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "65983a91a49c847937e790c61ab2c0d9569565e3d8461f59ceb421abe873e009", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "2e5782846e0d294aed397165fe58a9e752aa3074ecc760458b595e0645114bbc", "generation_seconds_batch": 26.796472709043883, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.0411396324634552, "target_z": -0.6132812495553993, "ce": 6.078212261199951, "genuineness": 4.268857479095459, "identity": 0.7142575979232788, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/65983a91a49c847937e790c61ab2c0d9569565e3d8461f59ceb421abe873e009.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": "65c8ce9875df5e4c6692ea5eec37270ef859a39c496a16b79428d45fb41ab557", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "964ec58989fc60772e2831ac23eb586ed86887806c7f1a056c424f7aff0ecb6f", "generation_seconds_batch": 17.443555210949853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.012047111988067627, "target_z": -0.5117187362173774, "ce": 6.356205940246582, "genuineness": 4.281409740447998, "identity": 0.36292409896850586, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter, bis zum kleinen Holztor. Ich warte auf der anderen Seite.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/65c8ce9875df5e4c6692ea5eec37270ef859a39c496a16b79428d45fb41ab557.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "660b24b21302972ca91acfe784d760a52c035fc69b5308dd9274eb323ff5f4f0", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "a57f8465797dc0cd6eaa84be4f32556a08adf843ae116912e5bb0f10c8cd028e", "generation_seconds_batch": 25.175372371915728, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1552421748638153, "target_z": -0.23828123532817594, "ce": 6.348263263702393, "genuineness": 4.243753433227539, "identity": 0.519760012626648, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/660b24b21302972ca91acfe784d760a52c035fc69b5308dd9274eb323ff5f4f0.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "661d64e740123f9fce686d5ee7060eccc4cda1baf203cd945c6285c5072249c2", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "daf0372e76314e7f85d13022f7614e3a8a5eac980e6db96c32e97039e96359b1", "generation_seconds_batch": 23.02163333108183, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.026366621255874634, "target_z": -0.48437498043756794, "ce": 6.1894097328186035, "genuineness": 4.061748504638672, "identity": 0.437298059463501, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/661d64e740123f9fce686d5ee7060eccc4cda1baf203cd945c6285c5072249c2.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "669704703967e2133c7f3ea230f6b26a4f7cfe1a8c79f71a6d9a81d50457bd80", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "bec3b813acbaafc6626492826fa52eea655e6856d5f6de8a4d9d9cde4710c651", "generation_seconds_batch": 22.526967811980285, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08773592114448547, "target_z": -0.3671875044460073, "ce": 6.109982967376709, "genuineness": 3.9048471450805664, "identity": 0.4802447557449341, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/669704703967e2133c7f3ea230f6b26a4f7cfe1a8c79f71a6d9a81d50457bd80.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "66d8050b7ed4a8c3e7b172bd8fdc7cbb295f3a759dcac9ce0bd46d7df5bd0094", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "66c398c42d48979bd7a3c7d0a3bdca32cba4eaa5029d49589fb6f2737ef72dd3", "generation_seconds_batch": 28.158178649959154, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.012500613927841187, "target_z": -0.5585937379957803, "ce": 6.046441555023193, "genuineness": 4.369274139404297, "identity": 0.4997527301311493, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana, vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/66d8050b7ed4a8c3e7b172bd8fdc7cbb295f3a759dcac9ce0bd46d7df5bd0094.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": "66f55f510220bd4ca81f8b591bbe3bba2443c18c9bce43f9956363a5ae04c3a8", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aed7bada87d5059face451d4e4c4d02059c18d3ea39917ca02f4d5b007dafb73", "generation_seconds_batch": 22.176203629001975, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04068611562252045, "target_z": -0.4570312531122051, "ce": 6.0305562019348145, "genuineness": 4.105680465698242, "identity": 0.7472100257873535, "hypothesis": "La route tourne à gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/66f55f510220bd4ca81f8b591bbe3bba2443c18c9bce43f9956363a5ae04c3a8.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "67d48a68ae1bc7ef511e31866f05d24eda5f9b5d3153af2fa44105223f7f2e04", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b315d2a81d5612cd0c6f03faf84277f3345dcfc7ab628ca998e9bf0901761a3c", "generation_seconds_batch": 21.664499387028627, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00022676587104797363, "target_z": -0.5351562655610255, "ce": 6.221180438995361, "genuineness": 4.068024158477783, "identity": 0.5843422412872314, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/67d48a68ae1bc7ef511e31866f05d24eda5f9b5d3153af2fa44105223f7f2e04.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "681c407917b580b8fe4dd3996daa52239841bc091e25fa927f6b66fc6fb9942d", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "68003caf480b295a75d4332d8d3567c131fe039109b8ad572a40a2361f91eeaa", "generation_seconds_batch": 23.317241381970234, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8732631206512451, "target_z": 1.1328124955539927, "ce": 5.871703147888184, "genuineness": 4.243753433227539, "identity": 0.528781533241272, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/681c407917b580b8fe4dd3996daa52239841bc091e25fa927f6b66fc6fb9942d.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": "68208da5656dfaa3d52987c359dd8895dec43c040916f441cf55dcd3b04949fd", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4c092e0418eab51736ec19160ff4b05d21bf4bde29e59a0e861dc10e3d0aef47", "generation_seconds_batch": 21.229308939073235, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.018184036016464233, "target_z": -0.5, "ce": 6.141753673553467, "genuineness": 4.281409740447998, "identity": 0.5776633620262146, "hypothesis": "La roue tourne à gauche après le pont, continue, jusqu'à la petite barrière en bois, je t'attendrai de l'autre côté.", "wer": 0.05, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/68208da5656dfaa3d52987c359dd8895dec43c040916f441cf55dcd3b04949fd.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "683f4aacd66bbf4fb1f55e7c3abfa5373498f31aa3573d71c47e814e7ea5c85e", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "d45ff5937e699243bb7bdc2c5daa0cb898870c6e8d812362ff7c69155ab939c0", "generation_seconds_batch": 25.650032219011337, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6748356819152832, "target_z": 0.753906254890608, "ce": 6.133810997009277, "genuineness": 3.9801597595214844, "identity": 0.12348702549934387, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/683f4aacd66bbf4fb1f55e7c3abfa5373498f31aa3573d71c47e814e7ea5c85e.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "68aea8e94f27e79ad8c314aabdd702ee63063b4c67082d750ec23073380dbede", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "bd97b278467aa86581353940fff72d95f4e0b02d94fba53bc62913d3767900c9", "generation_seconds_batch": 29.87791913608089, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.327396035194397, "target_z": 2.0, "ce": 6.252951145172119, "genuineness": 4.18099308013916, "identity": 0.5710351467132568, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/68aea8e94f27e79ad8c314aabdd702ee63063b4c67082d750ec23073380dbede.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "68e305720c6a8f9005222944ac95995cf8e234e4a5f53a89b076767693689b3a", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "ecf07640136669c8373671750e220a24f51ad42b85836132153da4dba0eb0ec6", "generation_seconds_batch": 17.171336216968484, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07034794986248016, "target_z": -0.40039062833450545, "ce": 5.597680568695068, "genuineness": 3.973883628845215, "identity": 0.4172777533531189, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/68e305720c6a8f9005222944ac95995cf8e234e4a5f53a89b076767693689b3a.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "691432503b4f0de63d3d8cae65051beab96da090ee160272dc0b8faa6bcbc7b7", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "30ac579e992302f0bb347e86c076753da8d737a557476d8b7b11ea360382aa2d", "generation_seconds_batch": 26.384779763990082, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10716953873634338, "target_z": -0.3300781256669011, "ce": 6.300607204437256, "genuineness": 3.7667744159698486, "identity": 0.41216373443603516, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/691432503b4f0de63d3d8cae65051beab96da090ee160272dc0b8faa6bcbc7b7.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "691e8a955832313d1b760558feaa3dc33089c4d2e779c02b31b2cabec4d4742e", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ab467205bcd9c00de63e91554dbf9098887b9597b5040c4318ab76db70f3719b", "generation_seconds_batch": 25.828170730965212, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4828649759292603, "target_z": 2.2968750302328496, "ce": 6.1894097328186035, "genuineness": 4.143336772918701, "identity": 0.5200291872024536, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/691e8a955832313d1b760558feaa3dc33089c4d2e779c02b31b2cabec4d4742e.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "6921ea9e3835743e5d8224b9fd2b758635c9748ba0e5a025efe35bdba48a8e91", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9abf7cca57d5c8cb42ae1eba45de5716635dcfde39a4fc93d75ef376ca3f2835", "generation_seconds_batch": 24.031754920026287, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00022676587104797363, "target_z": -0.5351562655610255, "ce": 6.149696350097656, "genuineness": 4.111956596374512, "identity": 0.5183535218238831, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/6921ea9e3835743e5d8224b9fd2b758635c9748ba0e5a025efe35bdba48a8e91.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "69490fbafececa33e54e67a9a46021edeb229518bc9a807f45c39167feba56d4", "bucket": "emo_Malevolence_Malice", "phase": "wording", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "73558a83905a34c3ac00205890ed33473f9fca2a8170c41df900037389d24c27", "generation_seconds_batch": 19.71595651691314, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06523384153842926, "target_z": -0.4101562513338022, "ce": 6.419747352600098, "genuineness": 4.444586753845215, "identity": 0.5503568649291992, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/69490fbafececa33e54e67a9a46021edeb229518bc9a807f45c39167feba56d4.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "69d550b418be2cfcfbadc650e4d06db15b1aea128e7df031b76a7a9dabaa680e", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "becd9060b5660aac437022b3fef93ea478cfc6977df5d88ade7d429a9a998bad", "generation_seconds_batch": 25.79413764702622, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09387286007404327, "target_z": -0.35546873977418325, "ce": 6.141753673553467, "genuineness": 4.20609712600708, "identity": 0.42692771553993225, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/69d550b418be2cfcfbadc650e4d06db15b1aea128e7df031b76a7a9dabaa680e.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "6a043e20a277686d14b2b1e1dfc804ac41ebd08a537a4a800a828b1c1fe02e15", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "2c1527ab4a3cb6997af34044de54356eac83f0982d9a5c76cda1a53d7917d3b2", "generation_seconds_batch": 27.606042865896598, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0696449279785156, "target_z": 1.507812509781216, "ce": 6.022613525390625, "genuineness": 4.143336772918701, "identity": 0.14753949642181396, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/6a043e20a277686d14b2b1e1dfc804ac41ebd08a537a4a800a828b1c1fe02e15.mp3" }, { "id": "heldout-en-4", "language": "en", "sentences": [ "I wrote the number on the envelope.", "Read it carefully... and check the last two digits.", "Call me when you have a moment." ], "text": "I wrote the number on the envelope. Read it carefully... and check the last two digits. Call me when you have a moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-4", "key": "6a61bc20d43b3e97640d8dd45682f7055c5dbaa5aa987cdb55f8a7d7a432a2c2", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I wrote the number on the envelope.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I wrote the number on the envelope.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Intense Malevolence Malice. 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": "c59ce4c35036d2a12fa2f5637b573d9e7f58c507e9269fbcf78977ba3a661b27", "generation_seconds_batch": 20.750103883910924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.026366621255874634, "target_z": -0.48437498043756794, "ce": 6.443574905395508, "genuineness": 4.093128204345703, "identity": 0.42650216817855835, "hypothesis": "I wrote the number on the envelope, read it carefully, and check the last two digits. Call me when you have a moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/6a61bc20d43b3e97640d8dd45682f7055c5dbaa5aa987cdb55f8a7d7a432a2c2.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "6abe83d2286e922cf9459429cdd1ef5164c46f80623f8ada1ebd6ac4b9688f20", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ab833e9152fc5fed6576e03ffe60ed45e0149145a0dfe62308b4a284ffdcafd2", "generation_seconds_batch": 20.82998289493844, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.026820123195648193, "target_z": -0.5859374937755898, "ce": 6.300607204437256, "genuineness": 4.356721878051758, "identity": 0.6207431554794312, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/6abe83d2286e922cf9459429cdd1ef5164c46f80623f8ada1ebd6ac4b9688f20.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "6ac04965a950c2464cdc97e118f40ad996fdd4e27db8bb5ab12a2babc670c544", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "5dedcdfaf5a6aa23e9a87ca6fab720313cabb70a881fded63de3d4fda816031a", "generation_seconds_batch": 26.76419600599911, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3633864223957062, "target_z": 0.15917969472476184, "ce": 6.133810997009277, "genuineness": 4.256305694580078, "identity": 0.5938131809234619, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/6ac04965a950c2464cdc97e118f40ad996fdd4e27db8bb5ab12a2babc670c544.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": "6adb65abfa414281cb7e885be2404ee525460240f01deb4a5a26c0aaae0086d8", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0087542c10a423a56faff5912fcda36abcfcc3ffa8233db6102d7c5e4e4bd1a4", "generation_seconds_batch": 21.229308939073235, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0059101879596710205, "target_z": -0.5234374724347548, "ce": 6.109982967376709, "genuineness": 4.381826400756836, "identity": 0.494361937046051, "hypothesis": "La route tourne à gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/6adb65abfa414281cb7e885be2404ee525460240f01deb4a5a26c0aaae0086d8.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "6af2f92705665209b84c0c01882e948dfec5cd4377feee22ebdb0938eb4a504b", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "456df081d612a4507effc3f1feed240e8f001a3df9bf24f2e395667146ab6329", "generation_seconds_batch": 23.02163333108183, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.022275328636169434, "target_z": -0.49218749021878394, "ce": 6.205295085906982, "genuineness": 4.036643981933594, "identity": 0.6254453659057617, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/6af2f92705665209b84c0c01882e948dfec5cd4377feee22ebdb0938eb4a504b.mp3" }, { "id": "heldout-es-3", "language": "es", "sentences": [ "Alguien ha movido la pequeña silla azul.", "Estaba aquí... junto a la estantería alta.", "La colocaremos después de comer." ], "text": "Alguien ha movido la pequeña silla azul. Estaba aquí... junto a la estantería alta. La colocaremos después de comer.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-3", "key": "6b973c750e3616c70f28663c29496a0c74bea4a87dff4bad65cadc5394b99131", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Intense Malevolence Malice. 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": "e75d3e1450a873dc85bad184ce179637b1bbd26a10a3bbe63da06be8732d910a", "generation_seconds_batch": 19.37162971298676, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.039663299918174744, "target_z": -0.4589843663302858, "ce": 6.252951145172119, "genuineness": 4.607763767242432, "identity": 0.7194164991378784, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí, junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/6b973c750e3616c70f28663c29496a0c74bea4a87dff4bad65cadc5394b99131.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "6c64d5329f13f0537b15c3e5a3e348e97f7e84b158f2cda18488efc4f350a83f", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "3c7bc934332ebe048d7ecdabb1aeeb3cfd44be0e5a283668368264a7496938ab", "generation_seconds_batch": 18.795395730994642, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.05750477313995361, "target_z": -0.6445312317713701, "ce": 6.403861999511719, "genuineness": 4.319066047668457, "identity": 0.34701865911483765, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/6c64d5329f13f0537b15c3e5a3e348e97f7e84b158f2cda18488efc4f350a83f.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "6ce0c4f623b4f3ed69d03fa6c06dc988f255c16cf75f5f3bbbca594464d84713", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "efef9fe9a21eb4fa34c914b2aa108a696f3b346fa0c5518afc46ef1044ad8246", "generation_seconds_batch": 24.149918518029153, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.026820123195648193, "target_z": -0.5859374937755898, "ce": 6.268836498260498, "genuineness": 4.432034492492676, "identity": 0.6758052706718445, "hypothesis": "En Concron que no viene la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.29411764705882354, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/6ce0c4f623b4f3ed69d03fa6c06dc988f255c16cf75f5f3bbbca594464d84713.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "6d016ede7d19e5a4ab1a5b3c2c30ef247a5f94efca65f3986e3643653f54af4b", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "f8b15358b8163fa9d1ff30f78beed9d18f836af2369938406fd426024f792aa0", "generation_seconds_batch": 20.826012525940314, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.055005624890327454, "target_z": -0.42968749733239564, "ce": 5.990842819213867, "genuineness": 4.356721878051758, "identity": 0.44068729877471924, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/6d016ede7d19e5a4ab1a5b3c2c30ef247a5f94efca65f3986e3643653f54af4b.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": "6d29f2fa5ed4a4aeec0409ad66eaadfa41bd9713c658856635954cdd61824459", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c3dbead121079ab2e39ad805dccd5b47367fccf7544d5616af8921a953f912e5", "generation_seconds_batch": 23.100694670923986, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.016138404607772827, "target_z": -0.5039062264361613, "ce": 6.284721851348877, "genuineness": 4.319066047668457, "identity": 0.6702438592910767, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/6d29f2fa5ed4a4aeec0409ad66eaadfa41bd9713c658856635954cdd61824459.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "6d3f86daee51c89d762e57b64a5ad45de6cceacfa030a071954ab8474fc238ca", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "4b3256a60f8a95e0cbdeef1fa83714e35e74bf7e34d295b178ae5c68dfeff286", "generation_seconds_batch": 26.77258615894243, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08159899711608887, "target_z": -0.3789062406633847, "ce": 6.1020402908325195, "genuineness": 3.653805732727051, "identity": 0.35296130180358887, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/6d3f86daee51c89d762e57b64a5ad45de6cceacfa030a071954ab8474fc238ca.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "6db02fa2a6647da30b6b484bc5c819c507360b70fbfc60453cbb7c88d7fece4b", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "95147f690946b14edadbbbc012f32901dfff1fef2cd0840411801b406c7c7030", "generation_seconds_batch": 25.093500136979856, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07034794986248016, "target_z": -0.40039062833450545, "ce": 6.0305562019348145, "genuineness": 4.042920112609863, "identity": 0.7768943309783936, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/6db02fa2a6647da30b6b484bc5c819c507360b70fbfc60453cbb7c88d7fece4b.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "6dcc0caa23dd81757e909c797b2b3f422df6296ca573f49e2718af48c34d81c6", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "494093ce0fc9af33e9a66644e81b42ec91152b5c82e166836c6d267407f0cea9", "generation_seconds_batch": 29.258169593987986, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06318819522857666, "target_z": -0.4140625062244102, "ce": 6.141753673553467, "genuineness": 4.18099308013916, "identity": 0.5647591948509216, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/6dcc0caa23dd81757e909c797b2b3f422df6296ca573f49e2718af48c34d81c6.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "6ddc8b58cda63f84cbb87e2c97baf1f2f58f38cda63338e4dea0f4571d727ce6", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "d2e9ac8441385c5f1932fff11e29d69e250bce87f09b1ae3a3cc33f1a0215f3d", "generation_seconds_batch": 29.034140834002756, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.010001450777053833, "target_z": -0.5156250195624321, "ce": 6.181467056274414, "genuineness": 4.143336772918701, "identity": 0.18139809370040894, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/6ddc8b58cda63f84cbb87e2c97baf1f2f58f38cda63338e4dea0f4571d727ce6.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "6df4bfdc3a6f98b27c4660da2a91865627faba57415e50c4d65fbc109c55e687", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9a3019701db2ff2be321eb9b174b3e9cf1e992e892cc17dd7033830065d92616", "generation_seconds_batch": 20.672593241906725, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.028865784406661987, "target_z": -0.5898437771206445, "ce": 6.070269584655762, "genuineness": 4.231201171875, "identity": 0.6334102153778076, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/6df4bfdc3a6f98b27c4660da2a91865627faba57415e50c4d65fbc109c55e687.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "6e7d8c0df3591205f2488759ab72e56fe5d65b74784f56f2e8b49b6630baec6f", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "41819e8129ef6dfc2b3a3191242f57c28937c6fe1be6b1b9f2730d84afac7d3d", "generation_seconds_batch": 27.352845833054744, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.023298144340515137, "target_z": -0.49023437700070327, "ce": 6.149696350097656, "genuineness": 4.049196243286133, "identity": 0.6118401885032654, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/6e7d8c0df3591205f2488759ab72e56fe5d65b74784f56f2e8b49b6630baec6f.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "6edf306a30e1926101a59dc2750ebb28d9d1216b330f81ce9d74d0a07ae27352", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "73641fa56d648309eff7b24454ec61cab975bcbffbe7dd0aa1dd9eda687bd712", "generation_seconds_batch": 27.24187764490489, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00636368989944458, "target_z": -0.5468750017784029, "ce": 6.133810997009277, "genuineness": 4.494794845581055, "identity": 0.6699764132499695, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen daar samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/6edf306a30e1926101a59dc2750ebb28d9d1216b330f81ce9d74d0a07ae27352.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": "6ee023e55505edbc00f9af49ac93dff3d058ae7c2964844e4da004d3b12662ef", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "12234c9190ac58e5eb9c040adb85ab2ff659fda59d5bb8b99429ac69991514cf", "generation_seconds_batch": 18.511353975045495, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04477740824222565, "target_z": -0.44921874333098905, "ce": 6.23706579208374, "genuineness": 4.105680465698242, "identity": 0.4976970851421356, "hypothesis": "Tu mensaje llegó después de que me fuera, lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/6ee023e55505edbc00f9af49ac93dff3d058ae7c2964844e4da004d3b12662ef.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "6f94898833ceeace1df9d215d0787d0e4a49a16d1879d786e37265d1fdb46db6", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "b593d1db50a1cb39023ab84bba619bd45982c43cbb0562591cc3daf07c89d9b0", "generation_seconds_batch": 26.971920004929416, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.025343775749206543, "target_z": -0.48632815056454193, "ce": 5.804190158843994, "genuineness": 4.2249250411987305, "identity": 0.4999793767929077, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/6f94898833ceeace1df9d215d0787d0e4a49a16d1879d786e37265d1fdb46db6.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "701fc995eeeec06a24b8245d8902b57fc16f282eba538d887a8746766bc3ecb7", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "844930f77cd5318ff9054a8069ba2394269cc19bb648c5f7187a1ceedc13ac5c", "generation_seconds_batch": 27.440195964998566, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11739775538444519, "target_z": -0.31054687966830763, "ce": 6.070269584655762, "genuineness": 3.8922953605651855, "identity": 0.36377397179603577, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/701fc995eeeec06a24b8245d8902b57fc16f282eba538d887a8746766bc3ecb7.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "703b2376983ae1f972bf7c28ee078a687fc954bdd2ec425d510348b7d58e019a", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "519ad0f7bf9c67a571d2dcae4480a6af759990d4b592d57c018ade9bf13e983b", "generation_seconds_batch": 25.692224894068204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.025343775749206543, "target_z": -0.48632815056454193, "ce": 6.173524379730225, "genuineness": 4.218649387359619, "identity": 0.697887659072876, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/703b2376983ae1f972bf7c28ee078a687fc954bdd2ec425d510348b7d58e019a.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "706a7e0f633168b9f16ebe7a804aad8eb05253e4c68192e9a182cdf84b293419", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "4b53ab91d29518b08db91302434e4bc28340f5dc91cd155320ee74c664255446", "generation_seconds_batch": 25.255807731067762, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.08205249905586243, "target_z": -0.691406233549773, "ce": 6.348263263702393, "genuineness": 4.356721878051758, "identity": 0.4507048726081848, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/706a7e0f633168b9f16ebe7a804aad8eb05253e4c68192e9a182cdf84b293419.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "70953481a9c2629788ee982985a0c193b877f7b77a1aac212a0183a1e74e443f", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0707676cdeb17da819d0616fcac8aae496563f797913fe7c9b74cf14a812e390", "generation_seconds_batch": 29.204729752033018, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1374709606170654, "target_z": 3.5468750017784028, "ce": 6.1894097328186035, "genuineness": 4.21237325668335, "identity": 0.45532065629959106, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/70953481a9c2629788ee982985a0c193b877f7b77a1aac212a0183a1e74e443f.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "70edd633a5f1b69e7c1d6ca6ec2cac36d9858607302f6e09923108b161ec4331", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "6f126d701cf49c57f51d0c15aa07d1599dab08e7f865405bb6d0a32432da2709", "generation_seconds_batch": 29.036034844932146, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03250354528427124, "target_z": -0.47265624422019054, "ce": 5.792276382446289, "genuineness": 3.8797430992126465, "identity": 0.4290737509727478, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/70edd633a5f1b69e7c1d6ca6ec2cac36d9858607302f6e09923108b161ec4331.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "71685a3a8705c3139719c533b0a5749c7c0e771384cbb651a751b8317f99ac54", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "3eba12e1b1c3a8841862f6ededb22f0326de23138cb082aebf0243bf091e3177", "generation_seconds_batch": 26.04662565898616, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0018188953399658203, "target_z": -0.5312499822159709, "ce": 6.1894097328186035, "genuineness": 4.400654315948486, "identity": 0.6472347974777222, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/71685a3a8705c3139719c533b0a5749c7c0e771384cbb651a751b8317f99ac54.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": "718d70d5a6effc749acedebd25628e51f77d60e19ad23a0efe63deb049186d94", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4c391faa8c0e5f63331f175e6d0392635c3f7faac6774c02eb222da5c6db5466", "generation_seconds_batch": 23.100694670923986, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.012047111988067627, "target_z": -0.5117187362173774, "ce": 6.316492557525635, "genuineness": 4.494794845581055, "identity": 0.6522076725959778, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí, junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/718d70d5a6effc749acedebd25628e51f77d60e19ad23a0efe63deb049186d94.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": "719fae3cd460f34630a1d62818a412ca21926a19272267d2e552a83a7d8128bf", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e5017240657dc6a25713f8f3cfb3dd0f8b08f350c0c0e76dd218af12629c1a7e", "generation_seconds_batch": 17.1304574070964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00227239727973938, "target_z": -0.5390624919971869, "ce": 6.117925643920898, "genuineness": 4.381826400756836, "identity": 0.6014835834503174, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/719fae3cd460f34630a1d62818a412ca21926a19272267d2e552a83a7d8128bf.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "71ce9a99c159c1596ee64c58e4adc3c74c144dea79858157187e3bb7e859c5fd", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "bb071c271c2c7ad32dd42e7893be080fd0a7cf89d022f54b28e5edbe963dc307", "generation_seconds_batch": 18.403548086993396, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.06364172697067261, "target_z": -0.6562500248976408, "ce": 6.41180419921875, "genuineness": 4.2249250411987305, "identity": 0.2400200515985489, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/71ce9a99c159c1596ee64c58e4adc3c74c144dea79858157187e3bb7e859c5fd.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "72141c683564465e52f09d131b5e4068a326952fcb57f945905d4e0b8ab14c8c", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "ff7f8645cf3103388fef7cac816988f7c6ea4e248548e2c89407ed98d434199c", "generation_seconds_batch": 26.45458048803266, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.025343775749206543, "target_z": -0.48632815056454193, "ce": 6.252951145172119, "genuineness": 4.632867813110352, "identity": 0.748779296875, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/72141c683564465e52f09d131b5e4068a326952fcb57f945905d4e0b8ab14c8c.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "72535d7665bfee9817e893cb11892005960bd6f9226262c01324d85cff94bc36", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "1d325d6fd2a799a07ed0ef17fbbafc1788f845816a4e74a0b29a6cad6dbbd755", "generation_seconds_batch": 24.424292281037197, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10614672303199768, "target_z": -0.33203123888498176, "ce": 5.685050010681152, "genuineness": 3.9801597595214844, "identity": 0.44890573620796204, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/72535d7665bfee9817e893cb11892005960bd6f9226262c01324d85cff94bc36.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "72794c1c623ea575762b335eb99c6f71834fbd67a82b676945ef7a6c664a4291", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "e8be253e1a3641062c94c36d10a731dacc872b0f2b4c999ad8b977ac8675e076", "generation_seconds_batch": 26.63033886998892, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.043185263872146606, "target_z": -0.6171874759915607, "ce": 6.284721851348877, "genuineness": 4.287685871124268, "identity": 0.4681535065174103, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/72794c1c623ea575762b335eb99c6f71834fbd67a82b676945ef7a6c664a4291.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "7336105bef03bb043c269273db6d2af76bf949e5821fa4d0f7d86744ecf864d4", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "9af3e37bb7b878e76d3381ef850300dc6de28845bf5a593576c2dde511219838", "generation_seconds_batch": 27.147066656965762, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.004318058490753174, "target_z": -0.5429687753422415, "ce": 6.0146708488464355, "genuineness": 4.099404335021973, "identity": 0.5513389706611633, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/7336105bef03bb043c269273db6d2af76bf949e5821fa4d0f7d86744ecf864d4.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "734c98b0805a6a8ae5759535e1ae0035436ebded1ec57cea29bc45b3f353880f", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "681ce14f0bba2891314ab7d17dc03bc9432cb4c5c3a7eb048707d64964b74917", "generation_seconds_batch": 25.02052228199318, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.022275328636169434, "target_z": -0.49218749021878394, "ce": 6.117925643920898, "genuineness": 4.344170093536377, "identity": 0.6259996891021729, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/734c98b0805a6a8ae5759535e1ae0035436ebded1ec57cea29bc45b3f353880f.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "735e5faca6dccde58e10b82916e89814477ee57719107a56e292842c86d1c855", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "8c47c20f4e10bfa1a8ac091ae9fbf6e3526ca43f1dfb610ccf16c52acd09be13", "generation_seconds_batch": 19.062127122073434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.014092743396759033, "target_z": -0.5078125097812161, "ce": 6.1020402908325195, "genuineness": 4.2751336097717285, "identity": 0.6933314800262451, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/735e5faca6dccde58e10b82916e89814477ee57719107a56e292842c86d1c855.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "735ee271542f865727acc6a40e9bf740568f72f57c72e9f00126b7fd9e02aecb", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "c3aa62e743edec0c0d811e5522be99757680c14d0af8fee788f592fc3a149c3d", "generation_seconds_batch": 26.83741046988871, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.012500613927841187, "target_z": -0.5585937379957803, "ce": 6.356205940246582, "genuineness": 4.13078498840332, "identity": 0.21739104390144348, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/735ee271542f865727acc6a40e9bf740568f72f57c72e9f00126b7fd9e02aecb.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "73752ffdc11b6bd604511141762c85b7aec737e54983631ef379a7b990d02263", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "6c22e38cc2fcb8ce3e6c34fa0cd1519f9f41808359eec045e4f5e93f6b6290db", "generation_seconds_batch": 27.62500850192737, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04068611562252045, "target_z": -0.4570312531122051, "ce": 6.268836498260498, "genuineness": 4.086852550506592, "identity": 0.7718759179115295, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/73752ffdc11b6bd604511141762c85b7aec737e54983631ef379a7b990d02263.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "7407488f94cf5417a70cd98f83a83cce12b6e75a02097dedb15c66f7967c184a", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "e5becb177e1d5819e95ac5621c3c128836eac3d8d1b7513959156d98069174aa", "generation_seconds_batch": 26.634423728100955, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.05750477313995361, "target_z": -0.6445312317713701, "ce": 6.300607204437256, "genuineness": 4.18726921081543, "identity": 0.524478554725647, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/7407488f94cf5417a70cd98f83a83cce12b6e75a02097dedb15c66f7967c184a.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "747012555e9ac8c71511da10e35267e5acb9e05f5af36f49b88ad9ac264bf13b", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "5cedf5c56b4ab2c62d967a2222bf723519dc60a02430631b709517f845854363", "generation_seconds_batch": 28.988100476097316, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0059101879596710205, "target_z": -0.5234374724347548, "ce": 6.292664527893066, "genuineness": 4.3504462242126465, "identity": 0.6767248511314392, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/747012555e9ac8c71511da10e35267e5acb9e05f5af36f49b88ad9ac264bf13b.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "7489626cb6b5864d025f375362d6ce1fb3a2521c6c5dc987e75924f8a5cf04f5", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "867403c8488a4c76d6ee2d6583cbc729111b813b44f1a44a69303ef7e4dedcfc", "generation_seconds_batch": 23.272989260964096, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.035002708435058594, "target_z": -0.6015625133380219, "ce": 6.4515180587768555, "genuineness": 4.099404335021973, "identity": 0.4551200866699219, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/7489626cb6b5864d025f375362d6ce1fb3a2521c6c5dc987e75924f8a5cf04f5.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": "74d986e269120ba865886160fc00841327801d87a4fdbe9527afa7bb8e7547eb", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "36081806f5c1661a972ed27f33ed6a79c301bc17db50af66221a38e60be1d904", "generation_seconds_batch": 19.704933063942008, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.052959978580474854, "target_z": -0.43359375222300367, "ce": 6.070269584655762, "genuineness": 4.149612903594971, "identity": 0.6685492992401123, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/74d986e269120ba865886160fc00841327801d87a4fdbe9527afa7bb8e7547eb.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "74f812d5983a361cbd90b26a7812640a3e6c6a9e36f4b0d96983269fb935930c", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "e3561df30d1ea4cefbeb66cd4eb2e81a36d5a877ff37813f029a237fe9ed7368", "generation_seconds_batch": 23.626195815973915, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.007955819368362427, "target_z": -0.5195312459985935, "ce": 6.070269584655762, "genuineness": 4.118232727050781, "identity": 0.5811771750450134, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/74f812d5983a361cbd90b26a7812640a3e6c6a9e36f4b0d96983269fb935930c.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": "7568d2d800f5e2668f5d8ce65018d7a460b6abbe665f9d26f698cc6283d4dfd7", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "62e430840db8e2942fa1f7652ead3aaa2dd42891aeaf79d7144c75420677e438", "generation_seconds_batch": 17.240327014005743, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6956119537353516, "target_z": 2.703125083584937, "ce": 6.268836498260498, "genuineness": 4.325342178344727, "identity": 0.44925349950790405, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/7568d2d800f5e2668f5d8ce65018d7a460b6abbe665f9d26f698cc6283d4dfd7.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "75d958749bc098bdf84a9b921d23cd5ffa23d5a5e8f3fe0e359414da34e42b9d", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "f9ee4ee161d0b1dfe463d711d97917f4afcd1f8d515e42d4be064210ac328480", "generation_seconds_batch": 24.468150334083475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.004318058490753174, "target_z": -0.5429687753422415, "ce": 6.038498878479004, "genuineness": 4.356721878051758, "identity": 0.6268079280853271, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/75d958749bc098bdf84a9b921d23cd5ffa23d5a5e8f3fe0e359414da34e42b9d.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "763bf8bc9f208b39b3189bc35652c18de6d87786bf041fc387e5e0394c6da824", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "c541a659d53f0a0d9d2991f426ab2faadb51161f10d77d8266210b567fb24af5", "generation_seconds_batch": 22.363845715066418, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.026366621255874634, "target_z": -0.48437498043756794, "ce": 6.078212261199951, "genuineness": 4.111956596374512, "identity": 0.5570946931838989, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/763bf8bc9f208b39b3189bc35652c18de6d87786bf041fc387e5e0394c6da824.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "764e67f93e9c7bf34c027c9529edaf88481a63c4ec9e8c3c414e698f4078875e", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "47b48f43a79626fa09db50f393871c7a1f8c5b4e6ac3e12b4de311eb075ccf06", "generation_seconds_batch": 29.258169593987986, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02125251293182373, "target_z": -0.49414060343686467, "ce": 6.23706579208374, "genuineness": 4.262581825256348, "identity": 0.6873086094856262, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/764e67f93e9c7bf34c027c9529edaf88481a63c4ec9e8c3c414e698f4078875e.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "76aadd613faa0cc9dc057f8f5a3851cf45e975c1fe7d39aff3da42c0d2367c32", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "b2f479cb8bd9eae87794b385c2f4ae7e67e32a45d6fd122a09e068ef2d290721", "generation_seconds_batch": 26.500971652916633, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.022275328636169434, "target_z": -0.49218749021878394, "ce": 5.9431867599487305, "genuineness": 4.13078498840332, "identity": 0.542130708694458, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/76aadd613faa0cc9dc057f8f5a3851cf45e975c1fe7d39aff3da42c0d2367c32.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "76c12a6b909a96b80d7a3ef9bc9d022a238c27af2c74f66e0cad052f88a09814", "bucket": "emo_Malevolence_Malice", "phase": "wording", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "eda9402c7820d9dd451cae6a286d3b606480fc9bae0caa2a830be0dc163a87be", "generation_seconds_batch": 19.104797292035073, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00022676587104797363, "target_z": -0.5351562655610255, "ce": 6.062326908111572, "genuineness": 4.218649387359619, "identity": 0.600080132484436, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/76c12a6b909a96b80d7a3ef9bc9d022a238c27af2c74f66e0cad052f88a09814.mp3" }, { "id": "heldout-fr-3", "language": "fr", "sentences": [ "Quelqu’un a déplacé la petite chaise bleue.", "Elle était ici... près de la grande bibliothèque.", "Nous la remettrons après le déjeuner." ], "text": "Quelqu’un a déplacé la petite chaise bleue. Elle était ici... près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-3", "key": "77962898fb95ddc0e92a153772191fe01482e40f98d2245158cb92652e8a93c1", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Intense Malevolence Malice. 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": "e1b8aa17c154d251f01e670daba77ca7ff16ca7cc7242cc6aea6038c7e0b6804", "generation_seconds_batch": 19.135938877006993, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01045498251914978, "target_z": -0.5546875115596189, "ce": 6.109982967376709, "genuineness": 4.7332844734191895, "identity": 0.7064787149429321, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue, elle était ici, près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/77962898fb95ddc0e92a153772191fe01482e40f98d2245158cb92652e8a93c1.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "7834944e63c78caa1e7c2ed8f2bd4aafd90c9d70cbe2af24d625eade0cb954e2", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "39daa0d6b9237d93cb2d3dd0e9698a03eb0f566a8f766335620b3e4b03c61952", "generation_seconds_batch": 29.06242644006852, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.057051271200180054, "target_z": -0.4257812424417876, "ce": 6.38797664642334, "genuineness": 4.086852550506592, "identity": 0.5946002006530762, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/7834944e63c78caa1e7c2ed8f2bd4aafd90c9d70cbe2af24d625eade0cb954e2.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "78bb897f4e782322d85d7a8cd103314a6fed7c4296355f0b2042aecc6dd30e05", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "9c1778002ae253d307f3be54b960b7401d23896ddbcad1c0c9ebb834423f77c3", "generation_seconds_batch": 27.307952026021667, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.38333144783973694, "target_z": 0.19726563011290837, "ce": 6.24500846862793, "genuineness": 4.2249250411987305, "identity": 0.4666801989078522, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/78bb897f4e782322d85d7a8cd103314a6fed7c4296355f0b2042aecc6dd30e05.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": "7906e36dfc86140688a8eab57ece7089f09337419884e61b3ba0023a2f4a5bee", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3278bbf261d5de2b4a785ab5d92043ab0557be8ef5dfc96fb5fbc5131cd865c9", "generation_seconds_batch": 24.735126861953177, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09182721376419067, "target_z": -0.3593749946647913, "ce": 6.165581703186035, "genuineness": 3.9927117824554443, "identity": 0.48043590784072876, "hypothesis": "Jew bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.043478260869565216, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/7906e36dfc86140688a8eab57ece7089f09337419884e61b3ba0023a2f4a5bee.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": "79101a77ce04afae0886ab5b1983c81225b9350867d6437794b79c964144a8c7", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4fc73401bd3b9e2c33d9c5d9851c12b8115cbca6e764633dd944553752b7942e", "generation_seconds_batch": 19.460877311998047, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0018188953399658203, "target_z": -0.5312499822159709, "ce": 6.133810997009277, "genuineness": 4.049196243286133, "identity": 0.6247203350067139, "hypothesis": "Ik schreef het nummer op de envelop. Lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/79101a77ce04afae0886ab5b1983c81225b9350867d6437794b79c964144a8c7.mp3" }, { "id": "heldout-nl-1", "language": "nl", "sentences": [ "De sleutels liggen op de keukentafel.", "Ik keek in de la... en daarna naast het raam.", "Neem ze mee als je weggaat." ], "text": "De sleutels liggen op de keukentafel. Ik keek in de la... en daarna naast het raam. Neem ze mee als je weggaat.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-1", "key": "794d3b9296e3c4be7c1e4182a2fa3cf095f7d593990763ce0f19c1f5e499a989", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Intense Malevolence Malice. 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": "9b8fb6892f0e16835ca361345562a9768ec24fa505ebe70c883b6e067aa47746", "generation_seconds_batch": 20.599867051001638, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0018188953399658203, "target_z": -0.5312499822159709, "ce": 6.117925643920898, "genuineness": 4.293961524963379, "identity": 0.7507228851318359, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/794d3b9296e3c4be7c1e4182a2fa3cf095f7d593990763ce0f19c1f5e499a989.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "79b362fe82e871e868d1a8155fbfb2d9966d9445ccf374fc4365954ee0d70d6c", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "2c228bbfde906630f0aca817f9b42ac8a7292b2caa5f08dabdb23fc4d5d207cf", "generation_seconds_batch": 27.190540881012566, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2293967604637146, "target_z": -0.09667967338392686, "ce": 5.601652145385742, "genuineness": 3.9425034523010254, "identity": 0.3605521321296692, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/79b362fe82e871e868d1a8155fbfb2d9966d9445ccf374fc4365954ee0d70d6c.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "7a0f19c9a79bc93d24ae1d6ae9021839beeaa8212605c7296b325f2c8fe86564", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "88d7fa7933c3a5560e2e60e8ff29dd416d59684b0407c8fa5b3f1dbabc6a355f", "generation_seconds_batch": 27.367997149005532, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.012047111988067627, "target_z": -0.5117187362173774, "ce": 6.125868320465088, "genuineness": 4.457139015197754, "identity": 0.5460332036018372, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/7a0f19c9a79bc93d24ae1d6ae9021839beeaa8212605c7296b325f2c8fe86564.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "7a607f7aa0cc6c31568f0d0322e6623507a7cc3d26f6e76fddaade1f94a0c29a", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "814163060747cf33f3d8b87560e46a6f536381622ece8c863f031b13fc5c23d6", "generation_seconds_batch": 27.873861138010398, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.685063898563385, "target_z": 0.7734375008892015, "ce": 6.332377910614014, "genuineness": 4.369274139404297, "identity": 0.639388382434845, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/7a607f7aa0cc6c31568f0d0322e6623507a7cc3d26f6e76fddaade1f94a0c29a.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": "7a94da23638cb508f02476fddfe5c663f6982ac6bed528181bcdbca166ba1ab4", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The keys are on the kitchen table.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5f2ec321b8642080fef654440aab02ee81f2118d10d60929b95babf00be94c8b", "generation_seconds_batch": 18.254465957987122, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0038645267486572266, "target_z": -0.5273437557798095, "ce": 6.38003396987915, "genuineness": 4.18099308013916, "identity": 0.4290136694908142, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you leave.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/7a94da23638cb508f02476fddfe5c663f6982ac6bed528181bcdbca166ba1ab4.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "7aa73263f5593dedc4cc3271156041ae377fb2da2f50125470af937d37aff337", "bucket": "emo_Malevolence_Malice", "phase": "wording", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "1ce24a7efa9a90dfcc37fe233e3482d6a1efe0ddcd5640eed97952d460eecb1c", "generation_seconds_batch": 19.71595651691314, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06523384153842926, "target_z": -0.4101562513338022, "ce": 6.395919322967529, "genuineness": 4.080576419830322, "identity": 0.3175305128097534, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/7aa73263f5593dedc4cc3271156041ae377fb2da2f50125470af937d37aff337.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "7ac4de5edbfe0cfafc3221a16afaf4bae5d1ad8cded75b0e9e7fe0b77d807125", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "a53a7b25f0a1f605fbd7641206bd7a022bab9e85a5c46c1ea52446fee1122277", "generation_seconds_batch": 24.48615619703196, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03045789897441864, "target_z": -0.47656249911079857, "ce": 6.332377910614014, "genuineness": 4.15588903427124, "identity": 0.518841028213501, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/7ac4de5edbfe0cfafc3221a16afaf4bae5d1ad8cded75b0e9e7fe0b77d807125.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "7b149f61e55849e009e6f3c5ab7fb1a820b203cb9a7188b10b7e30f924b7109c", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "4a1f7cab8b04e880ef75654690d05a08ca008d44ae184a243bd87fa8a87f9bc0", "generation_seconds_batch": 20.201530438032933, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03250354528427124, "target_z": -0.47265624422019054, "ce": 5.792276382446289, "genuineness": 4.093128204345703, "identity": 0.5168598890304565, "hypothesis": "La porte t'est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/7b149f61e55849e009e6f3c5ab7fb1a820b203cb9a7188b10b7e30f924b7109c.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": "7baf31db956bfea6fd835ff39aa3eca3dec7a7f919fb043163738cf694a510ba", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c4d491ab8001060ff06667763565a92c6351b523a90c3462cbf800695ade5bc4", "generation_seconds_batch": 17.22793620498851, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.016591906547546387, "target_z": -0.5664062477769963, "ce": 6.213237762451172, "genuineness": 4.369274139404297, "identity": 0.6473856568336487, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai regardé dans le tiroir, puis près de la fenêtre. Prends-les quand tu partiras.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/7baf31db956bfea6fd835ff39aa3eca3dec7a7f919fb043163738cf694a510ba.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "7bb7e82e2bb80978db334b3b263e9a65cc7b644184ee1d9904f8eeb7b48cc978", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "e20a736a3356d095c7bd43e438a3dae4fc3d36bba4194db1979d52775dca078a", "generation_seconds_batch": 23.911939401063137, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00636368989944458, "target_z": -0.5468750017784029, "ce": 6.125868320465088, "genuineness": 4.362998008728027, "identity": 0.6304497122764587, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/7bb7e82e2bb80978db334b3b263e9a65cc7b644184ee1d9904f8eeb7b48cc978.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "7bc0f538fcf4ed41409368d43b616b9dc0e889378036dc49ffdff16fc30decbe", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e161edd8ac90db9b27b421ef9cafdf50ae5e009cef2008563866af1c8161f9f8", "generation_seconds_batch": 27.757757390034385, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.016138404607772827, "target_z": -0.5039062264361613, "ce": 5.871703147888184, "genuineness": 4.042920112609863, "identity": 0.4842270612716675, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/7bc0f538fcf4ed41409368d43b616b9dc0e889378036dc49ffdff16fc30decbe.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "7c25ab854567564fe20d201ab05ce59676f4a682e2ec7a20a946f3d61ff2266c", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "c87def07ddbe868ca2c1565d85860c629bba775fd03961cbb9274bbc47d60b64", "generation_seconds_batch": 28.6198508720845, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.012047111988067627, "target_z": -0.5117187362173774, "ce": 5.7406487464904785, "genuineness": 3.9550557136535645, "identity": 0.4732476472854614, "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/emo_Malevolence_Malice/7c25ab854567564fe20d201ab05ce59676f4a682e2ec7a20a946f3d61ff2266c.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "7c86832d3bb84b9c4b7a13240bec4f249b027ad63a5512dc872d28004fcef238", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "1aa8ddc2c1a46a77c5e704a979b7ca220d4dcf4405260e6c2fd5db7840064f94", "generation_seconds_batch": 22.363845715066418, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0038645267486572266, "target_z": -0.5273437557798095, "ce": 6.316492557525635, "genuineness": 4.231201171875, "identity": 0.6288191080093384, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen heen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/7c86832d3bb84b9c4b7a13240bec4f249b027ad63a5512dc872d28004fcef238.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "7ca82b1aaec445a5e6fa02db1242b5b6bb37b6900c5a826197fb110c96a16099", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "90f33359913152e7ada9cab034dccda4facf7b27700129adf22d85f643a514c2", "generation_seconds_batch": 27.644933697069064, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5728732347488403, "target_z": 2.4687499039662426, "ce": 6.165581703186035, "genuineness": 3.923675537109375, "identity": 0.5074279308319092, "hypothesis": "Ik heb het kleine douche gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/7ca82b1aaec445a5e6fa02db1242b5b6bb37b6900c5a826197fb110c96a16099.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "7d6799a7b7f4d3b203f11a003ee6f9fe75e665ff293511aaccad25077ac1bc85", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "e012c27566757eb49dc3805e2b702695af0ba364e35748189e2c7521180ba515", "generation_seconds_batch": 23.626195815973915, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05398279428482056, "target_z": -0.43164063900492294, "ce": 6.157639026641846, "genuineness": 4.193545341491699, "identity": 0.4939693808555603, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/7d6799a7b7f4d3b203f11a003ee6f9fe75e665ff293511aaccad25077ac1bc85.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "7dd5785b95d1f7cf524717320c0d175e76f36be397ace2a5dc0080423efc5220", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "bd555d8882045a68cee1b7be765bf1e4c7c0feefed61572b857f3b0722fcf3f3", "generation_seconds_batch": 22.552532361936755, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.010001450777053833, "target_z": -0.5156250195624321, "ce": 6.205295085906982, "genuineness": 4.256305694580078, "identity": 0.658581554889679, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/7dd5785b95d1f7cf524717320c0d175e76f36be397ace2a5dc0080423efc5220.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "7e0a52827099a64f984fe0ab4e44bdaa74f3611c93f73e99ee3db52745e9b26c", "bucket": "emo_Malevolence_Malice", "phase": "wording", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "b86a7346c4d0f530581ac61a242f99c40454d950fe7577e713b926a5ce4b4ee2", "generation_seconds_batch": 20.429856966948137, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.061596065759658813, "target_z": -0.6523437415525861, "ce": 6.435632705688477, "genuineness": 4.231201171875, "identity": 0.2648766040802002, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/7e0a52827099a64f984fe0ab4e44bdaa74f3611c93f73e99ee3db52745e9b26c.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": "7e710cb1fab3091236fe58a6c224b6cabeddf496e175633465b4d34c1b201604", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a0a63411023c3a329c751b12d7bacacb411c0b0587bdda84569737eb44616b15", "generation_seconds_batch": 17.78761451295577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04580022394657135, "target_z": -0.4472656301129084, "ce": 6.078212261199951, "genuineness": 4.381826400756836, "identity": 0.5168179869651794, "hypothesis": "De wacht reis linksaf naar de brug. Loop verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.14285714285714285, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/7e710cb1fab3091236fe58a6c224b6cabeddf496e175633465b4d34c1b201604.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "7e714cf958501bc90ff503bf56f771fcb1c4aea95a71a96216a319e1fa824e64", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "f99c6bf6c8e4d06a58f8de01590ffd440270f5e86fa7d7500273f9bef83c55d4", "generation_seconds_batch": 29.113067654077895, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1378541886806488, "target_z": -0.27148438767112076, "ce": 6.133810997009277, "genuineness": 4.030367851257324, "identity": 0.6651936769485474, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/7e714cf958501bc90ff503bf56f771fcb1c4aea95a71a96216a319e1fa824e64.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "7eab44f4942c3dac23feb27a2a4c8234a7f1cdddad9b454825258dbaa220f7fc", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3920b5789fc531b1ee3977363ee935fb2e2a2ee31800b942626f0390b4c58a37", "generation_seconds_batch": 25.586506217019632, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02841225266456604, "target_z": -0.48046875400140654, "ce": 6.2767791748046875, "genuineness": 4.049196243286133, "identity": 0.4239909052848816, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/7eab44f4942c3dac23feb27a2a4c8234a7f1cdddad9b454825258dbaa220f7fc.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": "7f5e356734c427c15d4d508657ab077d958f357ce6ef954094d9b7fef8f6dac4", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "508b61128d9924ec141346dbac2de8f0b51c774b6697db6e4a437dcbacd40459", "generation_seconds_batch": 18.774774680961855, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2242826521396637, "target_z": -0.10644529638322357, "ce": 6.284721851348877, "genuineness": 4.15588903427124, "identity": 0.08416342735290527, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade, dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/7f5e356734c427c15d4d508657ab077d958f357ce6ef954094d9b7fef8f6dac4.mp3" }, { "id": "heldout-en-1", "language": "en", "sentences": [ "The keys are on the kitchen table.", "I checked the drawer... then looked beside the window.", "Take them with you when you leave." ], "text": "The keys are on the kitchen table. I checked the drawer... then looked beside the window. Take them with you when you leave.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-1", "key": "7f6fce1a6f733719f45a363f28fee4e24b2f07a3396aa07b42392913d7c24239", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The keys are on the kitchen table.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The keys are on the kitchen table.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Intense Malevolence Malice. 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": "4b7c6acc04144a6ab9062e7f9f104da201df5ef90ce120206bbe36833195d6cb", "generation_seconds_batch": 22.634793738950975, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.004318058490753174, "target_z": -0.5429687753422415, "ce": 6.356205940246582, "genuineness": 4.231201171875, "identity": 0.33631592988967896, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you leave.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/7f6fce1a6f733719f45a363f28fee4e24b2f07a3396aa07b42392913d7c24239.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "7f96cc2fe596656eb5eb8d757951141c4a900a8f26f990ae97f3b3ecce2c0c8b", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "3e2d1a9c996bd55bce88494ecfd29b79adf1e97433c5b7107ad867c859a6acb9", "generation_seconds_batch": 20.25811673491262, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.061596065759658813, "target_z": -0.6523437415525861, "ce": 6.38003396987915, "genuineness": 4.331617832183838, "identity": 0.16390696167945862, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/7f96cc2fe596656eb5eb8d757951141c4a900a8f26f990ae97f3b3ecce2c0c8b.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": "805857678fd0e99a3c01ef4b2773e2cc87c5b1c94fc89c0cedc95e69396730d2", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ffa19406e8cc1a9e88af24039f50ede1be21c42864e4ecb2f17b37df4d867f19", "generation_seconds_batch": 23.03110868507065, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.012047111988067627, "target_z": -0.5117187362173774, "ce": 6.117925643920898, "genuineness": 4.457139015197754, "identity": 0.6193759441375732, "hypothesis": "Las llaves están en la mesa de la cocina. Miren el cajón y luego junto a la ventana. Llévatelas cuando te vayas.", "wer": 0.08695652173913043, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/805857678fd0e99a3c01ef4b2773e2cc87c5b1c94fc89c0cedc95e69396730d2.mp3" }, { "id": "heldout-es-1", "language": "es", "sentences": [ "Las llaves están en la mesa de la cocina.", "Miré en el cajón... y luego junto a la ventana.", "Llévatelas cuando te vayas." ], "text": "Las llaves están en la mesa de la cocina. Miré en el cajón... y luego junto a la ventana. Llévatelas cuando te vayas.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-1", "key": "80c2597a6b8f43b5338a6b37d654868d679f46ded48514b42837ab18fba327ad", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Intense Malevolence Malice. 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": "15e7848d23f63a479bfae828ddb6c886059ca94eae593c3952e26081555028f8", "generation_seconds_batch": 22.309804894961417, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02125251293182373, "target_z": -0.49414060343686467, "ce": 6.300607204437256, "genuineness": 4.394378662109375, "identity": 0.5905731916427612, "hypothesis": "Las llaves están en la mesa de la cocina. Miré en el cajón y luego junto a la ventana. Llévatelas cuando te vayas.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/80c2597a6b8f43b5338a6b37d654868d679f46ded48514b42837ab18fba327ad.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "81148ff867d8635cd69b40b63a3d4cedc93c54475824d3c5c0f91a75dc9536ad", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "83f3a0ce8d9313ad89f3981d72853b625a691caf7a03061ed9accad6889a59ad", "generation_seconds_batch": 26.447590688010678, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03454919159412384, "target_z": -0.4687499893295825, "ce": 6.205295085906982, "genuineness": 3.873466968536377, "identity": 0.135498508810997, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/81148ff867d8635cd69b40b63a3d4cedc93c54475824d3c5c0f91a75dc9536ad.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "8197be7b002a751d459e7cf7805a1f29d51ee2500916624b60dad037d85ccec9", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "d52de00135b4bd704095e14ae83a1445df18d5928506d6170069605b76061b23", "generation_seconds_batch": 19.40704492200166, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03454919159412384, "target_z": -0.4687499893295825, "ce": 6.165581703186035, "genuineness": 4.507347106933594, "identity": 0.49813878536224365, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/8197be7b002a751d459e7cf7805a1f29d51ee2500916624b60dad037d85ccec9.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "81b8f9b69548ac38292765771d33be9174c7be00e9402aadcff1bac27643402f", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "75733f025b1d05fa3a772f7e234955fbd66f76a7f1b92c112b9062e9aaf40d9d", "generation_seconds_batch": 28.163135805050842, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02022966742515564, "target_z": -0.49609377356383866, "ce": 5.982900142669678, "genuineness": 4.105680465698242, "identity": 0.46612462401390076, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/81b8f9b69548ac38292765771d33be9174c7be00e9402aadcff1bac27643402f.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "82376c1179b46ad941fa9beb06a5c9f711b4fb4db680da91730d415e8b0f15e7", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "a6de7fffb4e65770c79b4d0977ca52d23299fad7d40cda981b249994d14af01e", "generation_seconds_batch": 26.971920004929416, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.029435068368911743, "target_z": -0.47851564078332587, "ce": 5.974957466125488, "genuineness": 4.400654315948486, "identity": 0.47923192381858826, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/82376c1179b46ad941fa9beb06a5c9f711b4fb4db680da91730d415e8b0f15e7.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "82d9577b1f103bda7b5513bad92f832f180ea85ae5b2a0f5d8a59fb148a74ba7", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a90c339c24f4fb42e053858875cdd3a4bfa233e56fd878201b475293907e89a1", "generation_seconds_batch": 27.98958996508736, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.18592682480812073, "target_z": -0.17968749733239564, "ce": 6.427689552307129, "genuineness": 4.074300289154053, "identity": 0.5890514850616455, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/82d9577b1f103bda7b5513bad92f832f180ea85ae5b2a0f5d8a59fb148a74ba7.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": "82fae955d67578de155f71d852437c8c25d38a714ec3a43527837ecc0e72a92f", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Escribí el número en el sobre.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9fa9fd5faf150c17e4c300a36830b897bce1e43ecb7827e8b916dd8b704784fc", "generation_seconds_batch": 19.404513677000068, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08773592114448547, "target_z": -0.3671875044460073, "ce": 5.982900142669678, "genuineness": 4.1998209953308105, "identity": 0.7257106304168701, "hypothesis": "Escribí el número en el sobre, léelo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/82fae955d67578de155f71d852437c8c25d38a714ec3a43527837ecc0e72a92f.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": "8308a27d344afd0a08de467dfeffd666a7d4b23660ad83c8d7525dc77fb635fa", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "665af7416e144f0e98d01aa787544092ff65c4ce168607e54165e871d577269a", "generation_seconds_batch": 27.323564617079683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.024774491786956787, "target_z": -0.5820312673394284, "ce": 6.24500846862793, "genuineness": 4.111956596374512, "identity": 0.4388963580131531, "hypothesis": "Die Schlüsse liegen auf dem Küchentisch, ich suchte in der Schublade, dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/8308a27d344afd0a08de467dfeffd666a7d4b23660ad83c8d7525dc77fb635fa.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "847e283c3c1ab9f06295f5e780108c80505d9ebc73f60b0f33198798255ac9bc", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "4e8b4e2dc39fe20a841ab2f342e7b8d38485bd47c212a3d75ebb88cf5d808123", "generation_seconds_batch": 20.201530438032933, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1245575100183487, "target_z": -0.2968750017784029, "ce": 6.0305562019348145, "genuineness": 4.3127899169921875, "identity": 0.4697674512863159, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/847e283c3c1ab9f06295f5e780108c80505d9ebc73f60b0f33198798255ac9bc.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "8503bd91ac42bdf52a9dbcff9387305c1c7442bd9c2f549f259ceedcff74a13f", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "80b95fe0030bc93e93afed017f322fd448945234a02f10e5b6b0d0b79c1f7664", "generation_seconds_batch": 27.154015577980317, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.05955043435096741, "target_z": -0.6484375151164248, "ce": 6.1020402908325195, "genuineness": 4.419482707977295, "identity": 0.674824595451355, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/8503bd91ac42bdf52a9dbcff9387305c1c7442bd9c2f549f259ceedcff74a13f.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "8557c191e70b8b00b97b0c614c82790aba227d5ea7328920b081c47a0c5f4f65", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1c77d5b2142097ff597e4d471b470b39ec77e6ce9b6ff1c58b4b016d51e10770", "generation_seconds_batch": 27.835458765970543, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.40787917375564575, "target_z": 0.2441406318913113, "ce": 6.395919322967529, "genuineness": 4.143336772918701, "identity": 0.5091143846511841, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/8557c191e70b8b00b97b0c614c82790aba227d5ea7328920b081c47a0c5f4f65.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "85667a4e097db854448ec20d98f5bae0c77b3ffd5839b12e99de774d26f379bb", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "4a6596319cb09c84e2dad9e32d095d528a8b3bea9b88555fa3bd9a4200e14424", "generation_seconds_batch": 26.423480157973245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05193716287612915, "target_z": -0.43554686544108434, "ce": 6.229123115539551, "genuineness": 4.394378662109375, "identity": 0.47726941108703613, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/85667a4e097db854448ec20d98f5bae0c77b3ffd5839b12e99de774d26f379bb.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "8584f054e275167beb09230d780d4fbfc94b48d8a841365ff658c41d22b490c0", "bucket": "emo_Malevolence_Malice", "phase": "wording", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "b5e102831688a78355b3981573d6c4658420ec627c4e4403de2b858b18d69e9d", "generation_seconds_batch": 19.298099645064212, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03659482300281525, "target_z": -0.46484376289342116, "ce": 6.070269584655762, "genuineness": 4.168440818786621, "identity": 0.7650558948516846, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/8584f054e275167beb09230d780d4fbfc94b48d8a841365ff658c41d22b490c0.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "85a975f991996f85fcb887bf08b291f69065a88bd928c4452ae5e1ea217f6098", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "0c386a86dea8227d67ddebbe58d1a45e7170765f83812b196116a9b7b0886c3b", "generation_seconds_batch": 20.6772598689422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00227239727973938, "target_z": -0.5390624919971869, "ce": 6.0305562019348145, "genuineness": 4.099404335021973, "identity": 0.5476677417755127, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/85a975f991996f85fcb887bf08b291f69065a88bd928c4452ae5e1ea217f6098.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "85fd9702568c631c77c1398582129632591fdc9ec5a0ebc8500d6c235320fa89", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "020702fe460caac8f53dd43b5b062205ab1d3d5c0554621e97cf051b3cce0434", "generation_seconds_batch": 19.419238707982004, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.035002708435058594, "target_z": -0.6015625133380219, "ce": 6.125868320465088, "genuineness": 4.256305694580078, "identity": 0.46748536825180054, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/85fd9702568c631c77c1398582129632591fdc9ec5a0ebc8500d6c235320fa89.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": "86910ff218a6030bb1954e3dfea3b03d62404e476eb751fb44957695dd6c2f61", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5bc5cc1ce9f1161061b8095446b9d2e62be3e24fc0ad801461411d35290acfbc", "generation_seconds_batch": 19.090801777900197, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08466745913028717, "target_z": -0.373046872554696, "ce": 6.141753673553467, "genuineness": 4.607763767242432, "identity": 0.67265784740448, "hypothesis": "J'ai écrit le numéro sur l'enveloppe. Lis-le bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/86910ff218a6030bb1954e3dfea3b03d62404e476eb751fb44957695dd6c2f61.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "869a0b7bb48ae6c2781ffff3a37672472463be42490d83b9fa83e400a794e833", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "56bba89d26ccf227a5adb09c17f20e939d044d81f0cf55bf6b9cee72178c51d7", "generation_seconds_batch": 17.27802036900539, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00227239727973938, "target_z": -0.5390624919971869, "ce": 6.141753673553467, "genuineness": 4.432034492492676, "identity": 0.524128794670105, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/869a0b7bb48ae6c2781ffff3a37672472463be42490d83b9fa83e400a794e833.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "8711b8f2cd24be38258fa8f1c8c5fccb01ea8b1378f99ae2b5033c01d3c524d3", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "7a769c7b69af23ee4d47e46723c1639c48d7a96fbb5bad3c82a935dc153bb70d", "generation_seconds_batch": 25.97593635006342, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09591849148273468, "target_z": -0.35156251333802185, "ce": 6.173524379730225, "genuineness": 4.394378662109375, "identity": 0.5525863170623779, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/8711b8f2cd24be38258fa8f1c8c5fccb01ea8b1378f99ae2b5033c01d3c524d3.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "877501a1fab4ea0602351ecb85231e86f3d3352e90c7b088901bfb1fbd016f21", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "fe3c381c4c2fe26994817caef83729dabb0158bc4115912d179cb0ba4645b6ac", "generation_seconds_batch": 18.949474229943007, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01045498251914978, "target_z": -0.5546875115596189, "ce": 5.911416053771973, "genuineness": 4.243753433227539, "identity": 0.6843435764312744, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/877501a1fab4ea0602351ecb85231e86f3d3352e90c7b088901bfb1fbd016f21.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "87b376dbf12fc8cad08282dbe608fd46cfad6155c663d09f0258867b993be9a9", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "bf6e478c116564125f10b75a0d79d728d18f801790e290ef10eafc32ef5eebb5", "generation_seconds_batch": 27.11878490296658, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.025343775749206543, "target_z": -0.48632815056454193, "ce": 6.197352409362793, "genuineness": 4.3881025314331055, "identity": 0.6914016008377075, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen daar samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/87b376dbf12fc8cad08282dbe608fd46cfad6155c663d09f0258867b993be9a9.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "87b6c9b1e23300050fe4ec6dd649794ad1a95874069daf43deae28378ab7a36f", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "c7b1ced7b16bc71cfcbee7218aef79e9c4c1d2bed975aed40dc052ea05b482a3", "generation_seconds_batch": 26.447590688010678, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.025343775749206543, "target_z": -0.48632815056454193, "ce": 6.292664527893066, "genuineness": 4.068024158477783, "identity": 0.15425032377243042, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/87b6c9b1e23300050fe4ec6dd649794ad1a95874069daf43deae28378ab7a36f.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "8823c873585034d05521563a7ba073f2d7ed08cefdfa086412529b1205fd14f7", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aaba39df7a1a2b1f420c34d7621295b9c809eda64f2a0775776b8fc6d7f36a21", "generation_seconds_batch": 25.077688689925708, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.007955819368362427, "target_z": -0.5195312459985935, "ce": 5.959072113037109, "genuineness": 4.055472373962402, "identity": 0.4371449053287506, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/8823c873585034d05521563a7ba073f2d7ed08cefdfa086412529b1205fd14f7.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "88822f12f9ef9208e6979f8c0ca81e4ca5f3b0b490249ad5c4a21d7a9c55069a", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4fc174f0e75adfd59d6af2a9fc875301b17438e2d0400bbf786e3525b858a485", "generation_seconds_batch": 25.967064597061835, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.05545914173126221, "target_z": -0.6406250053352087, "ce": 6.41180419921875, "genuineness": 4.262581825256348, "identity": 0.3825904428958893, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/88822f12f9ef9208e6979f8c0ca81e4ca5f3b0b490249ad5c4a21d7a9c55069a.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "892eb7d00c6e94b937d780a47fb8bbe7af1a51c534eb2298a81a5fc1284278e1", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "d65d3ebefcda24145d8c13f93eeea9654330f4f0aeaed3b387ecc47985aed598", "generation_seconds_batch": 27.715750394039787, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.022728830575942993, "target_z": -0.5781249839943737, "ce": 5.967014789581299, "genuineness": 4.193545341491699, "identity": 0.5978811979293823, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/892eb7d00c6e94b937d780a47fb8bbe7af1a51c534eb2298a81a5fc1284278e1.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": "89873abad228e7723d7e66ddd2c596068ca5f34fbe0c3b671d8bd91fb9978583", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I wrote the number on the envelope.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "17e624e32deac786c8863b569c0b2eb50218a2d5effcf9c994279039691e447b", "generation_seconds_batch": 18.92645937402267, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.057051271200180054, "target_z": -0.4257812424417876, "ce": 6.419747352600098, "genuineness": 3.8922953605651855, "identity": 0.15170475840568542, "hypothesis": "I wrote the number on the envelope. Read it carefully and check the last two digits. Call me when you have a moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/89873abad228e7723d7e66ddd2c596068ca5f34fbe0c3b671d8bd91fb9978583.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "8a3aafa448d4c1032b3f5379b76cdb525ef4d052266acae455e68acb1ef4d06b", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "2021e9eb911e81b5021083ef89fd80d492817f7166960bc45d2d5da2271d401b", "generation_seconds_batch": 25.36149588006083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02841225266456604, "target_z": -0.48046875400140654, "ce": 6.181467056274414, "genuineness": 4.381826400756836, "identity": 0.5270904302597046, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/8a3aafa448d4c1032b3f5379b76cdb525ef4d052266acae455e68acb1ef4d06b.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": "8b01c5b1ec479a14340da555996de5d41523b37f9bd98128f8f4d904946582a2", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "67521dff43e036d0e07bccab1d55e57787cd1672c9a8546e66e842fbac2ed95e", "generation_seconds_batch": 18.511353975045495, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3567380905151367, "target_z": 0.1464844018983441, "ce": 6.149696350097656, "genuineness": 4.243753433227539, "identity": 0.6822147965431213, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/8b01c5b1ec479a14340da555996de5d41523b37f9bd98128f8f4d904946582a2.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "8b2cb9526af70ccc2d46ba11be546051fdc83c76758f1b1acb83f4a7877bfcd3", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "e0f8feb39af8273611d7041ca60891ec4655820f142629371de9bb753dec6c4d", "generation_seconds_batch": 27.180471549043432, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5745991468429565, "target_z": 0.562500021340835, "ce": 6.1894097328186035, "genuineness": 4.174716949462891, "identity": 0.30992037057876587, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/8b2cb9526af70ccc2d46ba11be546051fdc83c76758f1b1acb83f4a7877bfcd3.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "8b61af73ebf28181195331ce74084783d96c2d8b62d7d33c23d41b02e1ab48e1", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "5dbfd7ef1a8179b06f85efc7a0f1314a9cbfd6a30e43e09ec85fd98731104bd5", "generation_seconds_batch": 30.506790620973334, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.055005624890327454, "target_z": -0.42968749733239564, "ce": 5.692992687225342, "genuineness": 3.9299514293670654, "identity": 0.646248459815979, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/8b61af73ebf28181195331ce74084783d96c2d8b62d7d33c23d41b02e1ab48e1.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "8ba0d208dfd9c93765926b78c00e86e01b009b915acf72bed7cb76f99a24b22f", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e342bddc113d85636597f6a1306d8b3741f2263c1ed43bb350da684e88181ca9", "generation_seconds_batch": 19.298494191025384, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1388770192861557, "target_z": -0.26953124599859346, "ce": 5.6413655281066895, "genuineness": 3.923675537109375, "identity": 0.6059706211090088, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/8ba0d208dfd9c93765926b78c00e86e01b009b915acf72bed7cb76f99a24b22f.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "8c28beb16eee9c66efcc83f8cca01b5077d00496dec0ba33d08ec3b2acc90590", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "381a85be7e40405803c4eb599102a6c3e4d7041aac87f0787ad565b132fce75f", "generation_seconds_batch": 23.616001727059484, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03659482300281525, "target_z": -0.46484376289342116, "ce": 6.022613525390625, "genuineness": 4.061748504638672, "identity": 0.3679617643356323, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/8c28beb16eee9c66efcc83f8cca01b5077d00496dec0ba33d08ec3b2acc90590.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": "8c2e615cfe554438d78ddfc86a0914c5864f994b7ec3139e8d6cd6f14e3763ce", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2ff51949f7e632ea73f8c25e4fe985a803216cb1256b3329035e0d8aaf951fa0", "generation_seconds_batch": 23.09301681793295, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.018184036016464233, "target_z": -0.5, "ce": 6.308549880981445, "genuineness": 4.243753433227539, "identity": 0.7579587697982788, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/8c2e615cfe554438d78ddfc86a0914c5864f994b7ec3139e8d6cd6f14e3763ce.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": "8c7d708d408fcfd835bb7ca95417f85943818c4e56c51b42626c923d96d61508", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ea4f6fbc72fe77be94af6864d3aa1eb459a8fdcf5a27efa31b131e83139dd78b", "generation_seconds_batch": 20.663879070896655, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1603562831878662, "target_z": -0.22851561232887924, "ce": 6.1894097328186035, "genuineness": 4.042920112609863, "identity": 0.3996622562408447, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam neem ze mee als je weggaat.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/8c7d708d408fcfd835bb7ca95417f85943818c4e56c51b42626c923d96d61508.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "8d3d4983591de329a8dfcf6e622a69dba4c3119105d99ed939e3156a6a68295c", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "aaf874d52996ce834334130aaab43f9d061b708b17e6a8b44c8b90563293ab3e", "generation_seconds_batch": 27.46585072297603, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02022966742515564, "target_z": -0.49609377356383866, "ce": 6.0146708488464355, "genuineness": 4.20609712600708, "identity": 0.6872820854187012, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/8d3d4983591de329a8dfcf6e622a69dba4c3119105d99ed939e3156a6a68295c.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": "8d71ac85ecebf47dfb2e999bbf55f0f85f75045f6fab889aff8adf39ec2cc079", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "291506353ea58189f6cffecea97c2ad711373edd5f64296ae316c2d35772af06", "generation_seconds_batch": 23.05466240493115, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01454627513885498, "target_z": -0.562500021340835, "ce": 6.229123115539551, "genuineness": 4.2249250411987305, "identity": 0.6593374609947205, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/8d71ac85ecebf47dfb2e999bbf55f0f85f75045f6fab889aff8adf39ec2cc079.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "8dad9239245cb9db0c57f4a76b0f6246233fd9800ede4437c4cc2e47954c21b4", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "321c40efcac1825d44eb8ee601d7912af4aa9d3e6c13b1749f2e3e7507939580", "generation_seconds_batch": 23.616001727059484, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.43549537658691406, "target_z": 0.29687503023284956, "ce": 5.982900142669678, "genuineness": 4.13078498840332, "identity": 0.5392218828201294, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/8dad9239245cb9db0c57f4a76b0f6246233fd9800ede4437c4cc2e47954c21b4.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": "8dfdf067ab04468092b4e22159b452a54b3c9e304ebeec2c206c8a4721b42e7a", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "952097c8a2a055abd08cf0695966555bb55400398c698a6f41426211119028c4", "generation_seconds_batch": 26.381959442049265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.42117583751678467, "target_z": 0.2695312175441468, "ce": 6.292664527893066, "genuineness": 3.860914945602417, "identity": 0.4681738615036011, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, ließ sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/8dfdf067ab04468092b4e22159b452a54b3c9e304ebeec2c206c8a4721b42e7a.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "8e36abf76f9354cba9df5d2bf44899e6e997b0594d5ba8345e88dc3ffcb6a246", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": 127, "seconds_generated": 10.16, "mp3_sha256": "189122457c6e06b8fbda79d9bf02a01069f5549ca7407f2250887b3622c45000", "generation_seconds_batch": 25.093174056033604, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0059101879596710205, "target_z": -0.5234374724347548, "ce": 6.213237762451172, "genuineness": 4.444586753845215, "identity": 0.7016118764877319, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/8e36abf76f9354cba9df5d2bf44899e6e997b0594d5ba8345e88dc3ffcb6a246.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "8ea13e8c118cf23f9ad2c8613da440384966dbaa6033d160304c823d6c202f5e", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "535829fb989c153433f3435569258a36680c584bb68218f46666f9b1e493058a", "generation_seconds_batch": 25.114715668954886, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.03704833984375, "target_z": -0.6054687397741833, "ce": 6.268836498260498, "genuineness": 3.9676074981689453, "identity": 0.6102129220962524, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/8ea13e8c118cf23f9ad2c8613da440384966dbaa6033d160304c823d6c202f5e.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "8ec1c65d7a9efcab780845b96a4c4bab9a06bf34c6ee9d0551a717bfd14146bf", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "bd60cefa0b90d720801dc86f8d5ba5d760a7b143bd1fb2be171b1b59884792d4", "generation_seconds_batch": 28.543067724909633, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.0493222177028656, "target_z": -0.6289062691178313, "ce": 6.332377910614014, "genuineness": 4.268857479095459, "identity": 0.4434412121772766, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/8ec1c65d7a9efcab780845b96a4c4bab9a06bf34c6ee9d0551a717bfd14146bf.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": "8f282cd065462b1badd80da42c149584fcb59f6ffd00cec8cbcb7fac8af43cca", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I wrote the number on the envelope.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d063784433c3df3de8533ce0c6f337121c3766de7b75b80b2b5399211fd26013", "generation_seconds_batch": 26.634888340020552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.059096917510032654, "target_z": -0.4218749875511796, "ce": 6.3641486167907715, "genuineness": 3.9173994064331055, "identity": 0.2581058442592621, "hypothesis": "I wrote the number on the envelope. Read it carefully, and check the last two digits. Call me when you have a moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/8f282cd065462b1badd80da42c149584fcb59f6ffd00cec8cbcb7fac8af43cca.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "8f381e1d3c6a1c5115295e33590e864086473fb2ec5ddf779e03c20a7e9408e9", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "33c32ad8f0847b33566494e390b55307e942564086513427d0a25485ea157dcd", "generation_seconds_batch": 26.026272892020643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.014092743396759033, "target_z": -0.5078125097812161, "ce": 6.38797664642334, "genuineness": 4.074300289154053, "identity": 0.18485337495803833, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/8f381e1d3c6a1c5115295e33590e864086473fb2ec5ddf779e03c20a7e9408e9.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "8f6b8bc9ba7d097f40481412f8faaff5f62c4387e843ed0637474261db5c2e96", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "45efd42b03723024410bc902eeb1deabe37ad1ab8a079ac4aa4c3c285a4c8941", "generation_seconds_batch": 20.342931573977694, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.039094001054763794, "target_z": -0.6093750231192379, "ce": 6.157639026641846, "genuineness": 4.287685871124268, "identity": 0.4673272967338562, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/8f6b8bc9ba7d097f40481412f8faaff5f62c4387e843ed0637474261db5c2e96.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "90ea2ac1b235b53f269a46a0faaa17a46df0d0d26b9b8901edcb42026a53ea58", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "055c02255c72832fca99b76fec1ae0c3cc974e757c2ae48eb5f71c541fdc1cb3", "generation_seconds_batch": 21.659862173954025, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.18643823266029358, "target_z": -0.17871094072335528, "ce": 6.23706579208374, "genuineness": 4.18099308013916, "identity": 0.5473300814628601, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/90ea2ac1b235b53f269a46a0faaa17a46df0d0d26b9b8901edcb42026a53ea58.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "90eeaa336de79fc289d248256a2a32d225650311a0c586c0c2c0f1f7dee14662", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "6bcb09fa041914b971d939634b60ebfd6127ccf714199c4ef3c5de7425082b43", "generation_seconds_batch": 19.361784846987575, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.06977865099906921, "target_z": -0.6679687611150182, "ce": 6.260893821716309, "genuineness": 4.086852550506592, "identity": 0.4137701094150543, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/90eeaa336de79fc289d248256a2a32d225650311a0c586c0c2c0f1f7dee14662.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "912e874a56da752f5fd75150e91d0b02f0b980a4eba734a35457e17354fe14a8", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "beb910f6f117e22338247cbff152c0214fbdf39a078e8712f0a2905ac7176a08", "generation_seconds_batch": 28.83332093397621, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01454627513885498, "target_z": -0.562500021340835, "ce": 6.078212261199951, "genuineness": 4.13078498840332, "identity": 0.46232858300209045, "hypothesis": "¡Gie! He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/912e874a56da752f5fd75150e91d0b02f0b980a4eba734a35457e17354fe14a8.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "914deac435c2a7b3402b3ac73a41e1c8705c56242e98ae88a081dfbcdf4c62bb", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "97ba831c02440872460b6124f9a246a31310a78900aa24d334feb9da32f6a761", "generation_seconds_batch": 18.716593179968186, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04886868596076965, "target_z": -0.4414062620042197, "ce": 5.903473377227783, "genuineness": 4.319066047668457, "identity": 0.18153829872608185, "hypothesis": "De deur staan nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/914deac435c2a7b3402b3ac73a41e1c8705c56242e98ae88a081dfbcdf4c62bb.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "917b45b57be5699d063746229dba0049363714a2bcf3dd094a9ca7cd2e5589b9", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "dce4d6d9a6ea4673794721d728d4e5b5304c099faaa78c7fb5264085ea01d749", "generation_seconds_batch": 26.634423728100955, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.07182428240776062, "target_z": -0.6718749875511796, "ce": 6.483288764953613, "genuineness": 4.250029563903809, "identity": 0.34228211641311646, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/917b45b57be5699d063746229dba0049363714a2bcf3dd094a9ca7cd2e5589b9.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "91a00fedd134b9cc0e69a80b58a6eda85186c676debdc69d7f22f7aa6f800df9", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "a927e4e074d308572655a71e6f7986900383f2dbb557a3d77716ac7be0d7da3f", "generation_seconds_batch": 28.102338333963417, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.028865784406661987, "target_z": -0.5898437771206445, "ce": 6.340320587158203, "genuineness": 4.268857479095459, "identity": 0.47822487354278564, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/91a00fedd134b9cc0e69a80b58a6eda85186c676debdc69d7f22f7aa6f800df9.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "91e47a78d7024ecf27dea3ff02b3c080546ca27f4881ae1b3ee10f30b7565a83", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "7b4c7ebbc124302375c2346cbd8ac73aa8c743d637912177c9a8a3d4fc9889fa", "generation_seconds_batch": 23.631674370961264, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07137078046798706, "target_z": -0.39843748666197815, "ce": 5.895530700683594, "genuineness": 4.005263805389404, "identity": 0.10464595258235931, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/91e47a78d7024ecf27dea3ff02b3c080546ca27f4881ae1b3ee10f30b7565a83.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "9226ce9ed6e62c00fae3d70eee866007db80a68056f1c2d2929ffdafbc54416c", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "7c94b2df0fe5fa4bb8539d3a9406d600fa686235dcb3712bdb12b8e783facb10", "generation_seconds_batch": 29.036034844932146, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.022275328636169434, "target_z": -0.49218749021878394, "ce": 6.09409761428833, "genuineness": 4.218649387359619, "identity": 0.583232045173645, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/9226ce9ed6e62c00fae3d70eee866007db80a68056f1c2d2929ffdafbc54416c.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "9228f043da83fc7885c19b119f17bf10c784b08b327195d400b20bbf189875b0", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "f64860534c4b6628ded8629f061e0cf91f7037b4fe6b311792d2f3a9c2b38a5d", "generation_seconds_batch": 24.969611271983013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05807408690452576, "target_z": -0.42382812922370694, "ce": 6.054384231567383, "genuineness": 4.143336772918701, "identity": 0.5811539888381958, "hypothesis": "ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/9228f043da83fc7885c19b119f17bf10c784b08b327195d400b20bbf189875b0.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "92dc6b1be715a4cb47b30d2752894ede5c3ff7cdeb82a3737daa18248a0b1dfa", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "2d9bb17df222f528f6ac6ad54772fb7981686c369540c42941934d7523337329", "generation_seconds_batch": 25.04809733503498, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10819235444068909, "target_z": -0.3281250124488204, "ce": 5.808161735534668, "genuineness": 3.998987913131714, "identity": 0.41122445464134216, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/92dc6b1be715a4cb47b30d2752894ede5c3ff7cdeb82a3737daa18248a0b1dfa.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "93c78fc09192cb0fae56e375e79797cf249e9c93a58d96fe93ef129002cfc448", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "cc1bf043fbb831f269a4ef25715ed317320049f3225282175c259222108511b9", "generation_seconds_batch": 25.097128701978363, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.701429009437561, "target_z": 0.804687426196279, "ce": 6.125868320465088, "genuineness": 4.042920112609863, "identity": 0.7986186742782593, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/93c78fc09192cb0fae56e375e79797cf249e9c93a58d96fe93ef129002cfc448.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "93fe2958729b3ae9f462a5bf59e515c9abc005417af82549bad954f89aa261da", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "559dbc1e56d783c28e0b19691051ded0fcabb9c93f9e470170b4ef0437438495", "generation_seconds_batch": 24.701448975945823, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.012500613927841187, "target_z": -0.5585937379957803, "ce": 6.125868320465088, "genuineness": 4.3504462242126465, "identity": 0.7765555381774902, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/93fe2958729b3ae9f462a5bf59e515c9abc005417af82549bad954f89aa261da.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "941e3476d271f73e4634c4d8ac4cff7c56414464779b7ec4b8a1fe962f685598", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "68ef2f3048fabdcfad00ef59f544e4bda98e4c7995757e018385add55d719ca4", "generation_seconds_batch": 26.69307792407926, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.010001450777053833, "target_z": -0.5156250195624321, "ce": 6.221180438995361, "genuineness": 4.570107460021973, "identity": 0.49862805008888245, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/941e3476d271f73e4634c4d8ac4cff7c56414464779b7ec4b8a1fe962f685598.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "94677bf092233d3ebbd547f58a6ef78f21d221a95dfc112dcf577c07d9d615f8", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "989ac826430f5ff9425f37de1964eeabd14302ac4bdf431f7517abab7b68b456", "generation_seconds_batch": 28.54341738205403, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.010001450777053833, "target_z": -0.5156250195624321, "ce": 6.125868320465088, "genuineness": 4.406930446624756, "identity": 0.752937376499176, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/94677bf092233d3ebbd547f58a6ef78f21d221a95dfc112dcf577c07d9d615f8.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": "955a6bb2470152b33f7116a23040a8be98436170f576b890baea208eb00d7924", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "637da6d7a8700bb93e91a8ba9ec79493ca386ae1959f602c288701edc465db28", "generation_seconds_batch": 18.774774680961855, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.008409321308135986, "target_z": -0.5507812282145643, "ce": 6.356205940246582, "genuineness": 4.369274139404297, "identity": 0.42818260192871094, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade, dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/955a6bb2470152b33f7116a23040a8be98436170f576b890baea208eb00d7924.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "956c7d2a2080bf0c5e3bb5a26d1d7a8261b1659124e9d5da078eac800c031edc", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "e2e76ce7cf1621351f8a7c003532c15760a2222daca637b39a5e9589814bc4b9", "generation_seconds_batch": 25.940985750989057, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.008409321308135986, "target_z": -0.5507812282145643, "ce": 5.6413655281066895, "genuineness": 3.9425034523010254, "identity": 0.6927132606506348, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/956c7d2a2080bf0c5e3bb5a26d1d7a8261b1659124e9d5da078eac800c031edc.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "95fac2b22a8f1ef2d9ed302068e4a1c107e8d26d6d9317a92b994059fc34c1c9", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "aea24f1d123449d4383546395da2ee3f52daedc90d49957cd5ac3c79c26ea533", "generation_seconds_batch": 25.91627170704305, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10512389242649078, "target_z": -0.3339843805575091, "ce": 6.24500846862793, "genuineness": 4.017816066741943, "identity": 0.21651440858840942, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/95fac2b22a8f1ef2d9ed302068e4a1c107e8d26d6d9317a92b994059fc34c1c9.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "96a32094c0bd591d06be82453afa0d0770d28c00eb497c8c48d92233188a0f96", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a1698eab64d9215730a41222972a23d88433bc04abb2951772316ed6d51399e0", "generation_seconds_batch": 26.47648220101837, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05602844059467316, "target_z": -0.42773438411431497, "ce": 6.2767791748046875, "genuineness": 4.293961524963379, "identity": 0.4980107247829437, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/96a32094c0bd591d06be82453afa0d0770d28c00eb497c8c48d92233188a0f96.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": "97a0a89aace328ba550ff26d3c13f3f76491fd9f31be45ed0bfd810c6ba7ac3f", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dd89ac7b1217edd3b191072f544d40d70e7b43d7dbeba765f80b939032256ae1", "generation_seconds_batch": 17.199707175022922, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01454627513885498, "target_z": -0.562500021340835, "ce": 6.205295085906982, "genuineness": 4.111956596374512, "identity": 0.7001148462295532, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue. Elle était ici, près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/97a0a89aace328ba550ff26d3c13f3f76491fd9f31be45ed0bfd810c6ba7ac3f.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "980e6d20507d723eddb66d84639d1baa3223199d143c1ca788c5db80fcd1a7e3", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "3205584418c17818fbfe04042fa63ff5b93cec671eceba25f4c176da3836546b", "generation_seconds_batch": 25.94337989005726, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.022275328636169434, "target_z": -0.49218749021878394, "ce": 6.300607204437256, "genuineness": 4.143336772918701, "identity": 0.13392677903175354, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/980e6d20507d723eddb66d84639d1baa3223199d143c1ca788c5db80fcd1a7e3.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "986406089e497a691d176764c70c4251cf36b5469204a4e1dfde4e8a65d1b6f1", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "f3a955a1a18c9f43f80608be5fbedd15fe00379b3d37b85a4ddd3a5c4046268f", "generation_seconds_batch": 25.041195454075933, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6380140781402588, "target_z": 0.683593723768557, "ce": 6.24500846862793, "genuineness": 4.21237325668335, "identity": 0.490063339471817, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschreiben.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/986406089e497a691d176764c70c4251cf36b5469204a4e1dfde4e8a65d1b6f1.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": "986c7c120b3f34295a107b7e254744151a28fd880a92a9330b87a2dc41ded6a1", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I wrote the number on the envelope.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4f84ebd4d000f2ae8b62a6f0f3846a58b1d6a9eb8c74d33e6b9d60a4fa76823e", "generation_seconds_batch": 26.634888340020552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04580022394657135, "target_z": -0.4472656301129084, "ce": 6.435632705688477, "genuineness": 4.105680465698242, "identity": 0.27120599150657654, "hypothesis": "I wrote the number on the envelope. Read it carefully, and check the last two digits. Call me when you have a moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/986c7c120b3f34295a107b7e254744151a28fd880a92a9330b87a2dc41ded6a1.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": "98d586c2546632142051f9f1408b9fbbaef702700aebc403441877cc5d526e85", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Escribí el número en el sobre.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c89dde973005073550de4932b2844168ab62ff1d9cd9af09a51921c1edcf84fc", "generation_seconds_batch": 23.035625742049888, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.014092743396759033, "target_z": -0.5078125097812161, "ce": 6.221180438995361, "genuineness": 4.168440818786621, "identity": 0.7332327961921692, "hypothesis": "Escribí el número en el sobre. Léelo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/98d586c2546632142051f9f1408b9fbbaef702700aebc403441877cc5d526e85.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "996c9ca96db7c6781d5d425c04752fd246dd071ba1f054946f0badd58ac6020f", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "b822ea245e804fd0118f24081447972db5b8f99bbe73696a4bcb25ba8924b567", "generation_seconds_batch": 27.147066656965762, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00636368989944458, "target_z": -0.5468750017784029, "ce": 6.0146708488464355, "genuineness": 4.243753433227539, "identity": 0.6908301115036011, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/996c9ca96db7c6781d5d425c04752fd246dd071ba1f054946f0badd58ac6020f.mp3" }, { "id": "heldout-de-0", "language": "de", "sentences": [ "Deine Nachricht kam, als ich schon weg war.", "Ich las sie zweimal... dann legte ich das Handy weg.", "Sag mir, wann wir reden können." ], "text": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal... dann legte ich das Handy weg. Sag mir, wann wir reden können.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-0", "key": "99bf1b7d3eed7b885fdae888ab00b4960791c3971d259b46f771306d2801e39b", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Intense Malevolence Malice. 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": "53486c9dd5168c3d5b4e8372e9f3a1d754d6e33bd187dc6d1128ecfa7ae2fd40", "generation_seconds_batch": 18.421020167996176, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.12148904800415039, "target_z": -0.3027343698870916, "ce": 6.483288764953613, "genuineness": 4.645420074462891, "identity": 0.44330382347106934, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/99bf1b7d3eed7b885fdae888ab00b4960791c3971d259b46f771306d2801e39b.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "9a614c94436bcf8d16e7c67ab596d9b00cc7721cc9e29e8c2981787aef21166b", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "dc0e4d0dd4f126cc1aed53acc583f54a0f7acbf7296e3a7a7c093dfd42a4fafc", "generation_seconds_batch": 24.957745958003215, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01454627513885498, "target_z": -0.562500021340835, "ce": 6.086154937744141, "genuineness": 4.632867813110352, "identity": 0.6279425621032715, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/9a614c94436bcf8d16e7c67ab596d9b00cc7721cc9e29e8c2981787aef21166b.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "9ab0537c65af3cfbddc493cccaabb8b7ce0578b0f4196c1256834cb7ba9fc791", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "a0a7914117cb6565769683db7eacf29ff6159d6bdef4f95df04ddacc2cf9a7aa", "generation_seconds_batch": 22.754951155977324, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03864046931266785, "target_z": -0.46093750800281313, "ce": 5.744620323181152, "genuineness": 4.086852550506592, "identity": 0.42829275131225586, "hypothesis": "La puerta sigue abierta, ¿eh? He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/9ab0537c65af3cfbddc493cccaabb8b7ce0578b0f4196c1256834cb7ba9fc791.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "9b8daf28bc7597195b378abf5d48af598d60ef70f52f838657dc1430af1a4343", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a0c36eb5c1c7210f04c2de60c6ecb352a0dff40ec928849408ebc2843a8af723", "generation_seconds_batch": 25.56417757703457, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02125251293182373, "target_z": -0.49414060343686467, "ce": 6.133810997009277, "genuineness": 4.074300289154053, "identity": 0.5091490745544434, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/9b8daf28bc7597195b378abf5d48af598d60ef70f52f838657dc1430af1a4343.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "9bb4c481be2d853116bd7b03cb98644e6d7e8ec98a1eafcfe5c03503d4d55ebb", "bucket": "emo_Malevolence_Malice", "phase": "wording", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "9a386ce15b2e6bb6d9c012677dab345aa282686eefed4b583911a4f6b872048f", "generation_seconds_batch": 20.27062622201629, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.022275328636169434, "target_z": -0.49218749021878394, "ce": 5.824047088623047, "genuineness": 4.086852550506592, "identity": 0.6097583770751953, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/9bb4c481be2d853116bd7b03cb98644e6d7e8ec98a1eafcfe5c03503d4d55ebb.mp3" }, { "id": "heldout-nl-3", "language": "nl", "sentences": [ "Iemand heeft het kleine blauwe stoeltje verplaatst.", "Het stond hier... naast de hoge boekenkast.", "We zetten het na de lunch terug." ], "text": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier... naast de hoge boekenkast. We zetten het na de lunch terug.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-3", "key": "9c09e940c5dab53287ffae2d1a8fa5296f41b2df71714656aaa2bb8830bf5162", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Intense Malevolence Malice. 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": "f0ea192364977708c3eb647f9eaae9387e6eb5987eff5417690e10498735a25d", "generation_seconds_batch": 19.02442735200748, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.014092743396759033, "target_z": -0.5078125097812161, "ce": 6.197352409362793, "genuineness": 4.444586753845215, "identity": 0.7554630041122437, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier, naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/9c09e940c5dab53287ffae2d1a8fa5296f41b2df71714656aaa2bb8830bf5162.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "9c7630f997344e415160f71c4067932303000ecc9f95a6cf32918c9cd0c88cb4", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "1bd7566699e790854a47f6fd533ed0ff36027c6a3334d3feff20200d2579bd46", "generation_seconds_batch": 30.35906647599768, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03250354528427124, "target_z": -0.47265624422019054, "ce": 6.348263263702393, "genuineness": 4.65797233581543, "identity": 0.6350365281105042, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/9c7630f997344e415160f71c4067932303000ecc9f95a6cf32918c9cd0c88cb4.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": "9c79631a27540e4ea714645aa399b84f30a50a666b59a73968f62a9b59d6577a", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b9fe7afeca71b2fc295a0e9aefa04335fd17b8597e835a35a630848f5099e620", "generation_seconds_batch": 22.92426662403159, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.26020926237106323, "target_z": -0.03784179623588645, "ce": 6.09409761428833, "genuineness": 3.9801597595214844, "identity": 0.38288962841033936, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/9c79631a27540e4ea714645aa399b84f30a50a666b59a73968f62a9b59d6577a.mp3" }, { "id": "heldout-es-2", "language": "es", "sentences": [ "La carretera gira a la izquierda tras el puente.", "Sigue... hasta la pequeña puerta de madera.", "Te esperaré al otro lado." ], "text": "La carretera gira a la izquierda tras el puente. Sigue... hasta la pequeña puerta de madera. Te esperaré al otro lado.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-2", "key": "9c969db20a6db380e08592f6aed9c9189b3db634d677ebe8f573f593ebdfd2d3", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Intense Malevolence Malice. 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": "858c3be2aabb0a0029813ffc92c1d2fa99bc55fae5341bf982f71750139ba5f6", "generation_seconds_batch": 20.12568565295078, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00022676587104797363, "target_z": -0.5351562655610255, "ce": 6.260893821716309, "genuineness": 4.268857479095459, "identity": 0.5944925546646118, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/9c969db20a6db380e08592f6aed9c9189b3db634d677ebe8f573f593ebdfd2d3.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": "9d0112a42b2b8c717b448b82ae4ca08cca6f3a38fdd875419b40f4c99cb53100", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4389119bbc56b4268aa0c0d6343603831fe16637484963ccfe2f2260084699de", "generation_seconds_batch": 23.03110868507065, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00022676587104797363, "target_z": -0.5351562655610255, "ce": 6.221180438995361, "genuineness": 4.111956596374512, "identity": 0.7527259588241577, "hypothesis": "Las llaves están en la mesa de la cocina. Miré en el cajón y luego junto a la ventana. Llévatelas cuando te vayas.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/9d0112a42b2b8c717b448b82ae4ca08cca6f3a38fdd875419b40f4c99cb53100.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "9d251982d9ff93cd8e91b2f8dea9bc71f2fc20f2d080b86bdb213d7c9f6f7bcd", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "f2f7df8f1f84f675cf65965c6ab0604ffb6a822cdc934b6f51560d174e783262", "generation_seconds_batch": 25.024099898990244, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00022676587104797363, "target_z": -0.5351562655610255, "ce": 6.213237762451172, "genuineness": 3.8420867919921875, "identity": 0.1580858826637268, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/9d251982d9ff93cd8e91b2f8dea9bc71f2fc20f2d080b86bdb213d7c9f6f7bcd.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "9db3811d32714992291f51013511dfe248473131ed2517e159e4831ac53895be", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "5ea4cef22616f516a915faeb44ddad902098d59860fbb20371c07475db4adf7d", "generation_seconds_batch": 26.641673012985848, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.010001450777053833, "target_z": -0.5156250195624321, "ce": 6.141753673553467, "genuineness": 4.557555198669434, "identity": 0.7272636890411377, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/9db3811d32714992291f51013511dfe248473131ed2517e159e4831ac53895be.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "9e19899e43987f809948eaba479521c9fa25ea51ce9360446adc378b24294931", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "126f0e8c2eeac1278bbde2f35078256fd20291f4b0f757865d1a8e7f99627e74", "generation_seconds_batch": 17.95443709997926, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00636368989944458, "target_z": -0.5468750017784029, "ce": 6.197352409362793, "genuineness": 4.024092197418213, "identity": 0.3025364279747009, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/9e19899e43987f809948eaba479521c9fa25ea51ce9360446adc378b24294931.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": "9e56b3d534dae19b1082a5add08d792065417de61ed28a0b095abf5e6681a9b7", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ef77430c39cc2c7dc216190d8d6fc3a58a673c48e1afe4cb4097fa6f7458d284", "generation_seconds_batch": 25.858792957034893, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03250354528427124, "target_z": -0.47265624422019054, "ce": 6.308549880981445, "genuineness": 4.607763767242432, "identity": 0.6115556955337524, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal. Dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/9e56b3d534dae19b1082a5add08d792065417de61ed28a0b095abf5e6681a9b7.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "9ed8513dab196401f7355a2100ed6df6b4e6d85f8a5f7b8007291a41e7e74f4a", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "95462d649993901a36a4f4a9c67bc90f1103067ff3063e75d2c8f578f2cf14ff", "generation_seconds_batch": 25.346991810016334, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10716953873634338, "target_z": -0.3300781256669011, "ce": 5.919358730316162, "genuineness": 4.068024158477783, "identity": 0.39360344409942627, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/9ed8513dab196401f7355a2100ed6df6b4e6d85f8a5f7b8007291a41e7e74f4a.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "9ee1ab12bfd5b6bffa86db5ace2059481e4d3413707b3c5331a1d01846110cd0", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "148d5d7d729559210f9eaef898bb6705f4e25a579e020f67a2b465e1fa11c339", "generation_seconds_batch": 22.66036220395472, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4395866394042969, "target_z": 0.3046874831051723, "ce": 5.839932441711426, "genuineness": 4.080576419830322, "identity": 0.5486103296279907, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/9ee1ab12bfd5b6bffa86db5ace2059481e4d3413707b3c5331a1d01846110cd0.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "9f5c52c2bde00a0157d095bcab4a314a06c0a24ece02dfdcfdc368f3eb2e50d1", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "29fe6289bbbea25cea752dda1c6e9e6c730d13a12ca25714be61fb28237f3c21", "generation_seconds_batch": 28.979571270989254, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11228364706039429, "target_z": -0.32031250266760436, "ce": 6.09409761428833, "genuineness": 3.9299514293670654, "identity": 0.25992175936698914, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/9f5c52c2bde00a0157d095bcab4a314a06c0a24ece02dfdcfdc368f3eb2e50d1.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "9f7014efb05a1b5f2a33900a63bd6b418617167d45043d3e0ed1b709bae4abc0", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "1020c7fa0fffbd1a02c84d876b274132ad9742e0da9c5388f96d6dc2f7811416", "generation_seconds_batch": 29.637260803952813, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.018184036016464233, "target_z": -0.5, "ce": 6.157639026641846, "genuineness": 4.2249250411987305, "identity": 0.6772603988647461, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/9f7014efb05a1b5f2a33900a63bd6b418617167d45043d3e0ed1b709bae4abc0.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "9fa29b8c75822d030d4378139de8989ee6805b8014f2a6156de52a3ec7ee56f1", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "905716022a73c3c1eca8dc5f3d349a02c769ed2d7e650d8b69b51efb2e3a61b6", "generation_seconds_batch": 25.825370609993115, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01045498251914978, "target_z": -0.5546875115596189, "ce": 6.2767791748046875, "genuineness": 3.998987913131714, "identity": 0.16030512750148773, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/9fa29b8c75822d030d4378139de8989ee6805b8014f2a6156de52a3ec7ee56f1.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "9fbdbcc17c31abaf386c58b4de34715b31f89fc005a166dcf091c6086044a1ff", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "e8f084d0b40244590a2d403c1149439cdeb09ea3b49d729a00835af8b4ea0b10", "generation_seconds_batch": 21.664499387028627, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06932513415813446, "target_z": -0.4023437415525862, "ce": 6.141753673553467, "genuineness": 4.080576419830322, "identity": 0.61805260181427, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/9fbdbcc17c31abaf386c58b4de34715b31f89fc005a166dcf091c6086044a1ff.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "9fdf7586b8f3368541967234cda9b6b37a468baf57bf8a63bf3e505e23fc394f", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e72e169095bdfbf35105b4724cdb2a279d77f3766d4b3955c3890d2e5f8a2559", "generation_seconds_batch": 25.828170730965212, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2304195761680603, "target_z": -0.09472656016584617, "ce": 6.054384231567383, "genuineness": 3.961331605911255, "identity": 0.6405410170555115, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/9fdf7586b8f3368541967234cda9b6b37a468baf57bf8a63bf3e505e23fc394f.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": "9fe1a4b60276a1050dce6bfb2ce9c7ec49e79fcf7505c66afba1acf5585dc215", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The road turns left behind the bridge.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8c0a791b3f8a7091c25d2a3b06b028ffdd535f7628ba4bce27e67ddb16041836", "generation_seconds_batch": 17.17746460007038, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.022728830575942993, "target_z": -0.5781249839943737, "ce": 6.316492557525635, "genuineness": 3.9550557136535645, "identity": 0.45042115449905396, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/9fe1a4b60276a1050dce6bfb2ce9c7ec49e79fcf7505c66afba1acf5585dc215.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "9ff07b597bf11b018709d85e51d5afb15ed0508e8983aa72df53f07f624bf917", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "e3fff5d720cc2538ce8fe3ac4d1962d1f0918bc05e16a13b7e3093f33924f013", "generation_seconds_batch": 28.158178649959154, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.03704833984375, "target_z": -0.6054687397741833, "ce": 6.284721851348877, "genuineness": 4.344170093536377, "identity": 0.6950746774673462, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/9ff07b597bf11b018709d85e51d5afb15ed0508e8983aa72df53f07f624bf917.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "a002b56b1ceb3348a13cc41ad052f013c63989c17b32666ab8e35629260305d1", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "713cb15a35b09062651693d23f0d546eb235487367f7e0248bf158dec8ef9c21", "generation_seconds_batch": 25.863272108021192, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4436779320240021, "target_z": 0.31249999288638836, "ce": 6.1894097328186035, "genuineness": 4.005263805389404, "identity": 0.12504422664642334, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/a002b56b1ceb3348a13cc41ad052f013c63989c17b32666ab8e35629260305d1.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "a083825197633b1c4b9244598c404308366b3e87e36b8aa04f49be1bc0e99e02", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "779fc3657cb6febaab869eb3b89c7aa7af8d4014118163b10155b3d3ba0f5fd6", "generation_seconds_batch": 24.19068988203071, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.0411396324634552, "target_z": -0.6132812495553993, "ce": 6.443574905395508, "genuineness": 4.356721878051758, "identity": 0.2043645977973938, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/a083825197633b1c4b9244598c404308366b3e87e36b8aa04f49be1bc0e99e02.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "a0cc2b8a283de9e2b8cef8b90e710548977aa22e206d4230bb32e85132ad2e11", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "983321628d5d83132ad07ab06210470f07d22023a19bc3811e1b2da3ceaa040c", "generation_seconds_batch": 18.69529991107993, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03659482300281525, "target_z": -0.46484376289342116, "ce": 6.213237762451172, "genuineness": 3.9801597595214844, "identity": 0.27512770891189575, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/a0cc2b8a283de9e2b8cef8b90e710548977aa22e206d4230bb32e85132ad2e11.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "a1014afcf4ea9ed815fcc15836c6b00e15909d478fedb6daaf103b5f35864282", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "9ec2939ce1a12774b2cc8c9fba7cf8b5ebaea77a132e1222dc0d250bc80a93fa", "generation_seconds_batch": 19.698353365063667, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04068611562252045, "target_z": -0.4570312531122051, "ce": 6.427689552307129, "genuineness": 4.105680465698242, "identity": 0.5204870104789734, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/a1014afcf4ea9ed815fcc15836c6b00e15909d478fedb6daaf103b5f35864282.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "a196cee39638dd322c3474bb33c00e82e99666cb6740ce65674db6b07f8ec960", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "1db3517e2c8777139b5b04723b19e5ac134227ca78c8d251aaa31a32c1a10549", "generation_seconds_batch": 24.468150334083475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.057051271200180054, "target_z": -0.4257812424417876, "ce": 5.831989288330078, "genuineness": 3.873466968536377, "identity": 0.35380691289901733, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/a196cee39638dd322c3474bb33c00e82e99666cb6740ce65674db6b07f8ec960.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "a226ab413170dd57edea819096d1a4d18549c63da06f0b0595104c9f973cfc15", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "e524879806ae9356eeb9018b43aa61e3484572737e8d64ef1008ef99f1eaaac5", "generation_seconds_batch": 24.969611271983013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.023298144340515137, "target_z": -0.49023437700070327, "ce": 6.157639026641846, "genuineness": 4.174716949462891, "identity": 0.6405389308929443, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/a226ab413170dd57edea819096d1a4d18549c63da06f0b0595104c9f973cfc15.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "a2631515f678dcb0ffbb95cbd9b9e2a10968123be1aac4555fdd64f160f006cb", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "01e4ffb71bcd29f309bb8e554f1d7427593b9da49a39ded6f8d3414881573664", "generation_seconds_batch": 26.253196183010004, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1266031563282013, "target_z": -0.2929687468877949, "ce": 6.038498878479004, "genuineness": 4.174716949462891, "identity": 0.5810719728469849, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/a2631515f678dcb0ffbb95cbd9b9e2a10968123be1aac4555fdd64f160f006cb.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "a26b64c6422a9b280b44dab5bb43451d3b4c95c30d61358d498406f028164675", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "e2a7afbcee066ba09c21aeef3ba9af9fb39207ad485ead93e392e684e9ed7b00", "generation_seconds_batch": 27.62500850192737, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03250354528427124, "target_z": -0.47265624422019054, "ce": 6.070269584655762, "genuineness": 4.143336772918701, "identity": 0.5111410617828369, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/a26b64c6422a9b280b44dab5bb43451d3b4c95c30d61358d498406f028164675.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "a2a3b9e0c785df52c22e790cafbe5f2e9b0c0088ab8f9dc95cb0b41694975560", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "62b731fe5bddd229d814c876e732b2de745e3b50cd3bb5df66812529039e3142", "generation_seconds_batch": 25.06534181407187, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01045498251914978, "target_z": -0.5546875115596189, "ce": 6.284721851348877, "genuineness": 4.419482707977295, "identity": 0.6929140686988831, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/a2a3b9e0c785df52c22e790cafbe5f2e9b0c0088ab8f9dc95cb0b41694975560.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "a355dbc84b3bd72fee1d959857146060ce23f15181990d346f105d2ef38cf119", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "e94327be53bbf77c6e77627fa6a2e25e92ee0c8e3de02b66c7fad4dfa4dbcac5", "generation_seconds_batch": 27.164152734098025, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5705078840255737, "target_z": 0.5546875684685123, "ce": 5.681078910827637, "genuineness": 3.948779582977295, "identity": 0.33594903349876404, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/a355dbc84b3bd72fee1d959857146060ce23f15181990d346f105d2ef38cf119.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "a40a23eda359dbda6c118b3437ee3815cc59827996b25c1ebf3eb55f7215770b", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "70d8b62b75e068c3f5c4784aac7b24d9bbc451697c01fc6edb341ed7f0ca815e", "generation_seconds_batch": 26.605277815950103, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.012500613927841187, "target_z": -0.5585937379957803, "ce": 6.1894097328186035, "genuineness": 4.570107460021973, "identity": 0.7740553617477417, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/a40a23eda359dbda6c118b3437ee3815cc59827996b25c1ebf3eb55f7215770b.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "a464f0b8c95479ddb5731da8266526d20ad6f0a5786b5b782ab4245172cd6cc6", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "911ad41884c88f42a041abb571a90596887a3ec9b8bb9975a8f63cf60c4665a2", "generation_seconds_batch": 26.251836268929765, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06011973321437836, "target_z": -0.4199218743330989, "ce": 6.38003396987915, "genuineness": 4.030367851257324, "identity": 0.4494086802005768, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/a464f0b8c95479ddb5731da8266526d20ad6f0a5786b5b782ab4245172cd6cc6.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "a4f291e894e1405831f8a119718d3f08aebc5c670f58af4b270b27998fdfbd0c", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "148e9ff79c8a28dafbeb664e17f85870af2edc0b7de4d5a9bfd719172f4d4cea", "generation_seconds_batch": 18.862858982989565, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2651955187320709, "target_z": -0.028320312388849817, "ce": 6.181467056274414, "genuineness": 3.948779582977295, "identity": 0.5086772441864014, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/a4f291e894e1405831f8a119718d3f08aebc5c670f58af4b270b27998fdfbd0c.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "a53a920156376e461ef10682d1dd95173e2bc23fdb8b4a08de30a4b51064e7eb", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "daeb7a44f613d8cf7c5bc9942f5864f995539693f6e7e235e461c93c6722bb49", "generation_seconds_batch": 20.675008430029266, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.007955819368362427, "target_z": -0.5195312459985935, "ce": 6.046441555023193, "genuineness": 3.8169827461242676, "identity": 0.5125756859779358, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/a53a920156376e461ef10682d1dd95173e2bc23fdb8b4a08de30a4b51064e7eb.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "a5d327e8f0e860016383d36d059288157bffb3995ac9dbf328590e902715c295", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d036bd8042f39680747ae277287590672f338e564ea8e65008c16cad5dcfe96c", "generation_seconds_batch": 25.495932615944184, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.40276506543159485, "target_z": 0.23437500889201457, "ce": 6.165581703186035, "genuineness": 3.9550557136535645, "identity": 0.7163921594619751, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/a5d327e8f0e860016383d36d059288157bffb3995ac9dbf328590e902715c295.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "a5db3a2ed46b951cafdd5b0b5a54ca18201c0163b656e9835b690430cbc4b548", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "de27bbd2398f989d4645aed7fa9df344466770cdbf488ce136daa60aa0d18881", "generation_seconds_batch": 24.701448975945823, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.0411396324634552, "target_z": -0.6132812495553993, "ce": 5.998785495758057, "genuineness": 4.068024158477783, "identity": 0.4236741065979004, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/a5db3a2ed46b951cafdd5b0b5a54ca18201c0163b656e9835b690430cbc4b548.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "a5f82df998bb0983a1281bf13d49e46320969e760bdfbf71f728d2a3ba2b34f1", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "07c5739e8340c4d0c69a887fe306a6bba349050c773a916e6550ec56478b0fcc", "generation_seconds_batch": 26.883039213018492, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0018188953399658203, "target_z": -0.5312499822159709, "ce": 5.895530700683594, "genuineness": 4.18099308013916, "identity": 0.5067723393440247, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/a5f82df998bb0983a1281bf13d49e46320969e760bdfbf71f728d2a3ba2b34f1.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "a6017b881a040fe8374567c18e46fd46a6b8b74502da70db3049144c96f6f5cd", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "4c8f5607fd8fe0020e16955a318fbd18e68e8a1998106af5b7cfe6c856bb46b6", "generation_seconds_batch": 20.465618073008955, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03250354528427124, "target_z": -0.47265624422019054, "ce": 6.070269584655762, "genuineness": 4.218649387359619, "identity": 0.6632530689239502, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/a6017b881a040fe8374567c18e46fd46a6b8b74502da70db3049144c96f6f5cd.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "a68411955ada5cae0ab7f2f22ffac734de7db3364c2ff9335923503ace72d562", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "161c318371cfa3ff8ebdfe2357c7d4f3dfef4c5139b9357b9f56ab5d82edade3", "generation_seconds_batch": 27.730090899043716, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.020683199167251587, "target_z": -0.5742187575582124, "ce": 6.133810997009277, "genuineness": 4.6203155517578125, "identity": 0.6505484580993652, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/a68411955ada5cae0ab7f2f22ffac734de7db3364c2ff9335923503ace72d562.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": "a704a4718e4ad015dc074c055248dde0c7bbd9777086e83f0ddebf0ca2c1edac", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a06b5c4312792b98639f4773849f4fa5d66346242ca650a5fa555c6f901424fb", "generation_seconds_batch": 25.603216346004046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.014092743396759033, "target_z": -0.5078125097812161, "ce": 6.316492557525635, "genuineness": 4.444586753845215, "identity": 0.5629773139953613, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/a704a4718e4ad015dc074c055248dde0c7bbd9777086e83f0ddebf0ca2c1edac.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "a70933098c73f0e697250f531b60d8c8a109860cd0301728cc73628532c9dc04", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "4a659a87b3bfb8aadc30d67b28b9f4ce86a8aa41eed70ba6c7c2da1a8c37a8cd", "generation_seconds_batch": 19.406625418108888, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.024774491786956787, "target_z": -0.5820312673394284, "ce": 6.205295085906982, "genuineness": 3.936227321624756, "identity": 0.15540571510791779, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/a70933098c73f0e697250f531b60d8c8a109860cd0301728cc73628532c9dc04.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "a8420f81635eb13aa65d1eeb942c0357975f1471240ac79946185b9db1819a4d", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "93b08738e9734196fb1576d6a1e6a16081a1257d925938072f2a4704f222ce5d", "generation_seconds_batch": 26.994403230957687, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.18132412433624268, "target_z": -0.188476563722652, "ce": 6.372091293334961, "genuineness": 4.118232727050781, "identity": 0.23063316941261292, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/a8420f81635eb13aa65d1eeb942c0357975f1471240ac79946185b9db1819a4d.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "a843be3aacb3beaaa2e9194a8821c372d25a3ba403f9418cc8c879d7f79978de", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "10e81f8819fc32f0f058cf7dcb09d5c71aaaf792adbc7fe78a3fff03a1c377f9", "generation_seconds_batch": 22.9304855678929, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.04727655649185181, "target_z": -0.6249999857727767, "ce": 6.284721851348877, "genuineness": 4.118232727050781, "identity": 0.5125145316123962, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/a843be3aacb3beaaa2e9194a8821c372d25a3ba403f9418cc8c879d7f79978de.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "a858eb2cac113d05faec5565f40d74344f092f206c0245885535298ef9cc9f03", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "919447a2c1f9540689bccf207da4e2b90b4afb4f7d095636889f9bb2bd4ff7e4", "generation_seconds_batch": 26.973854048992507, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11126083135604858, "target_z": -0.32226561588568503, "ce": 5.911416053771973, "genuineness": 4.017816066741943, "identity": 0.6126882433891296, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/a858eb2cac113d05faec5565f40d74344f092f206c0245885535298ef9cc9f03.mp3" }, { "id": "heldout-nl-4", "language": "nl", "sentences": [ "Ik schreef het nummer op de envelop.", "Lees het goed... en controleer de laatste twee cijfers.", "Bel me wanneer je even tijd hebt." ], "text": "Ik schreef het nummer op de envelop. Lees het goed... en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-4", "key": "a8bd001b1fcafd2f49b65d73e6ed12567e4c9664909a4d08700c9febfd977540", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Intense Malevolence Malice. 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": "ce9f26b6252ef3143c6dd127fc5f5ad8081570d8fdf1ed0749c1a8d7c642b10e", "generation_seconds_batch": 17.005645935074426, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01045498251914978, "target_z": -0.5546875115596189, "ce": 6.181467056274414, "genuineness": 4.105680465698242, "identity": 0.6154841184616089, "hypothesis": "Ik schreef het nummer op de envelop Lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/a8bd001b1fcafd2f49b65d73e6ed12567e4c9664909a4d08700c9febfd977540.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": "a8ce01877f2d96730b3ebd7400270addc9318aa04b6be248147da4272bf61cdc", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f082479ba488c4b095ca4a1ad078642305b957d5f64e4c608eaa1b17a534576b", "generation_seconds_batch": 18.432049656985328, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05091433227062225, "target_z": -0.43750000711361164, "ce": 6.268836498260498, "genuineness": 4.432034492492676, "identity": 0.6373111009597778, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier, neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/a8ce01877f2d96730b3ebd7400270addc9318aa04b6be248147da4272bf61cdc.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": "a93d500acac29ba19b016f7c48b07be2dd342f6143617ced3a673d505becc61d", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "00fdf56e26bb065508fa6ad4117e57cb3953c39cc25a2f85934eb2a5247e65e7", "generation_seconds_batch": 19.922058008960448, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.010001450777053833, "target_z": -0.5156250195624321, "ce": 6.38003396987915, "genuineness": 4.231201171875, "identity": 0.6509422063827515, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/a93d500acac29ba19b016f7c48b07be2dd342f6143617ced3a673d505becc61d.mp3" }, { "id": "heldout-fr-0", "language": "fr", "sentences": [ "Ton message est arrivé après mon départ.", "Je l’ai relu... puis j’ai rangé mon téléphone.", "Dis-moi quand nous pourrons parler." ], "text": "Ton message est arrivé après mon départ. Je l’ai relu... puis j’ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-0", "key": "a94b7d5da78ff217f8e5a4f7ae4520900bef0ad764d0e4ee898e69de10c28aea", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Intense Malevolence Malice. 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": "a9a9ca78d85c6994e4da6e79d9fe577643a680d47e3b65f0b285b671ce31cfa4", "generation_seconds_batch": 19.520779552985914, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02841225266456604, "target_z": -0.48046875400140654, "ce": 6.292664527893066, "genuineness": 4.482243061065674, "identity": 0.6178900003433228, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/a94b7d5da78ff217f8e5a4f7ae4520900bef0ad764d0e4ee898e69de10c28aea.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "a98f4175316cb99a6d43efe075480c4b645a9a60b3bffab25fa7b313b70ee5e7", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "ac4ddbe9c30fab6e11501b52730d38af6cae51425ed42726323fd975a86e3497", "generation_seconds_batch": 27.59941558202263, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.019206851720809937, "target_z": -0.49804688678191933, "ce": 6.229123115539551, "genuineness": 4.105680465698242, "identity": 0.5193650722503662, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/a98f4175316cb99a6d43efe075480c4b645a9a60b3bffab25fa7b313b70ee5e7.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "aa4497435ff924eadaaaa6c9d8bc4d1f9c243ec101813d509bdf18614a0804a9", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "fc5c73d35eeebd7e1f56dd95c245b368f81ee03bae905edf3fb9ad55e70db525", "generation_seconds_batch": 26.19028902600985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09591849148273468, "target_z": -0.35156251333802185, "ce": 6.340320587158203, "genuineness": 4.005263805389404, "identity": 0.3941690921783447, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/aa4497435ff924eadaaaa6c9d8bc4d1f9c243ec101813d509bdf18614a0804a9.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "aa6a554ed15e0bb15f8d9c374a05183effde39f700c58b90320840ed5a353da0", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "b081a1a0301e459ca21c17cfa56d14964117c480013df3806a4868d99b7914ac", "generation_seconds_batch": 28.573859137948602, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.0329570472240448, "target_z": -0.5976562299929672, "ce": 6.229123115539551, "genuineness": 4.413206577301025, "identity": 0.5950665473937988, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/aa6a554ed15e0bb15f8d9c374a05183effde39f700c58b90320840ed5a353da0.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": "aae26293178ffad233f0e9d681dacfaa90f63e7c5234751d0259c264c9b06215", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I wrote the number on the envelope.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d059a2acff3c59c3926e88e0d502d45d41fd21b4a81309168c635b75d2c023e3", "generation_seconds_batch": 17.349562702002004, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02841225266456604, "target_z": -0.48046875400140654, "ce": 6.260893821716309, "genuineness": 3.6663577556610107, "identity": 0.3401567041873932, "hypothesis": "I wrote the number on the envelope. Read it carefully and check the last two digits. Call me when you have a moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/aae26293178ffad233f0e9d681dacfaa90f63e7c5234751d0259c264c9b06215.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "aae77f283e30f478df266a95ae856fc1735725a95568cd27a129ec748211db5b", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "51d10766b37d86817412765844a9d398c67d11f0ca0a9cd4ec7be45d8f8f6149", "generation_seconds_batch": 20.142297467100434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.018184036016464233, "target_z": -0.5, "ce": 6.284721851348877, "genuineness": 4.432034492492676, "identity": 0.7480961084365845, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/aae77f283e30f478df266a95ae856fc1735725a95568cd27a129ec748211db5b.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "ab04a16a93afac4eef3d8579020132549300bbcb86f38e899a0acb0f83e04a57", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "4f6ecc66b24182f8413815f3fe88f62ae0e369fadd638881bfdd9836dfbbf14e", "generation_seconds_batch": 27.154015577980317, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.026820123195648193, "target_z": -0.5859374937755898, "ce": 6.117925643920898, "genuineness": 4.256305694580078, "identity": 0.6730715036392212, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ab04a16a93afac4eef3d8579020132549300bbcb86f38e899a0acb0f83e04a57.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "ab0d4055a66d3f1ee4ab4e3e04654a6fc05ca619dfed8820c96764011a129d7a", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "a3b03527add4b2e3d0a94aded007f9da096b18cada78e560b93d4c5fc696d3dd", "generation_seconds_batch": 27.17870296095498, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.014092743396759033, "target_z": -0.5078125097812161, "ce": 5.895530700683594, "genuineness": 4.243753433227539, "identity": 0.4872578978538513, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ab0d4055a66d3f1ee4ab4e3e04654a6fc05ca619dfed8820c96764011a129d7a.mp3" }, { "id": "heldout-en-0", "language": "en", "sentences": [ "Your message arrived after I had left.", "I read it twice... before putting the phone away.", "Let me know when we can talk." ], "text": "Your message arrived after I had left. I read it twice... before putting the phone away. Let me know when we can talk.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-0", "key": "ab7a94299f968552a7b966fa3391724b652d8e20483888a7767196f29a7f4a69", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Your message arrived after I had left.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Your message arrived after I had left.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Intense Malevolence Malice. 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": "886ce09804e8d794c13c398b1374863707916c7291ecac7cba6e1e345947b0fd", "generation_seconds_batch": 20.21946391707752, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02125251293182373, "target_z": -0.49414060343686467, "ce": 6.499174118041992, "genuineness": 4.174716949462891, "identity": 0.41498124599456787, "hypothesis": "Your message arrived after I had left. I read it twice, before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ab7a94299f968552a7b966fa3391724b652d8e20483888a7767196f29a7f4a69.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "ab81a6667e67397db6a0fff3a6d8395c9995285de02861f612671682a7daea7c", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "77a2547446a582bb0d0b368e4a9bed9f29418d3724143ead3430e4d91d1e8a1d", "generation_seconds_batch": 25.659666858031414, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.018184036016464233, "target_z": -0.5, "ce": 6.070269584655762, "genuineness": 4.457139015197754, "identity": 0.5734509229660034, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ab81a6667e67397db6a0fff3a6d8395c9995285de02861f612671682a7daea7c.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "ac438dcddfeb752bf8632d32728abe33d496f9478b5b3e2194abcd1a04fe3eba", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "5696947998a311c2b68302dbdbd1c213d487ca11545737638179bb9a86550463", "generation_seconds_batch": 25.957190423971042, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03864046931266785, "target_z": -0.46093750800281313, "ce": 6.229123115539551, "genuineness": 4.256305694580078, "identity": 0.6563022136688232, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ac438dcddfeb752bf8632d32728abe33d496f9478b5b3e2194abcd1a04fe3eba.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": "acaa79c27351ab46cee587d50d1198e9ac4090397c98ee04cf7baa3bf393091f", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The road turns left behind the bridge.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f77ba5821889c957b8ff6f6519dd682ece9b62a7731f9da5b680bc393d8a36ea", "generation_seconds_batch": 27.128824665094726, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01045498251914978, "target_z": -0.5546875115596189, "ce": 6.372091293334961, "genuineness": 3.8546390533447266, "identity": 0.4515281915664673, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/acaa79c27351ab46cee587d50d1198e9ac4090397c98ee04cf7baa3bf393091f.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "ad10bfb4c0705b40403247cda6b060e407722610c0cb72bd6ebc9ae00a264ee5", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "74f1861d43e71d457840b5cc8f77252dcd51f987b54321200d17eebe0087f1e0", "generation_seconds_batch": 25.241973585099913, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.004318058490753174, "target_z": -0.5429687753422415, "ce": 6.197352409362793, "genuineness": 3.8295347690582275, "identity": 0.2343919575214386, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ad10bfb4c0705b40403247cda6b060e407722610c0cb72bd6ebc9ae00a264ee5.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "ad4296de502bc4539c0337bc1228ef3d24922699477110014d4f58c60ab4505e", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "c8cf1aea019bedf99ed7c8b0c74768695b923454cb693d83e2a8ac94fe8cb89e", "generation_seconds_batch": 19.936476087081246, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1337629109621048, "target_z": -0.27929686899789014, "ce": 5.661221981048584, "genuineness": 4.030367851257324, "identity": 0.4733412563800812, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ad4296de502bc4539c0337bc1228ef3d24922699477110014d4f58c60ab4505e.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "ad8d66060e60dc90e2bab68f1a326590df68f9564c6d16384a401a961507f6b9", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "6eb2269a34ea7bfc283229bc5c0887cb836d2d8d4c12ae3f49a26d2a3993d962", "generation_seconds_batch": 26.32377220503986, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.010001450777053833, "target_z": -0.5156250195624321, "ce": 6.0146708488464355, "genuineness": 4.030367851257324, "identity": 0.7222225666046143, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ad8d66060e60dc90e2bab68f1a326590df68f9564c6d16384a401a961507f6b9.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "adc60fea369f46eb4267a664df1c71f32015f174caa25fab7dea56a50e0f59cf", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "e53ddcc1df5adeee06c7f18eae6a891bd04c1b8005547f8f34508293b23b1f61", "generation_seconds_batch": 21.216272560996003, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.012500613927841187, "target_z": -0.5585937379957803, "ce": 6.221180438995361, "genuineness": 4.457139015197754, "identity": 0.6751524209976196, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/adc60fea369f46eb4267a664df1c71f32015f174caa25fab7dea56a50e0f59cf.mp3" }, { "id": "heldout-nl-0", "language": "nl", "sentences": [ "Je bericht kwam nadat ik vertrokken was.", "Ik las het twee keer... en legde mijn telefoon weg.", "Zeg maar wanneer we kunnen praten." ], "text": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer... en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-0", "key": "adcc0121d6c11de6953f8243f84ad890832f75283c96342c1db246fddc443438", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Intense Malevolence Malice. 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": "6b661aa1fbf035b1fb04b5a13701add83c51adff5905f84ea42948d4024a2737", "generation_seconds_batch": 22.45649806095753, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.027389436960220337, "target_z": -0.48242186721948727, "ce": 6.260893821716309, "genuineness": 4.356721878051758, "identity": 0.5857939720153809, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/adcc0121d6c11de6953f8243f84ad890832f75283c96342c1db246fddc443438.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "adddce7e5df726b1e22e213e1e69c4b55cc70a97e43413a7d77cebb4abec5632", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "82ee494f4338c6f775b739792dd080d5d3ecb413d84d61d5096e8fb29193768a", "generation_seconds_batch": 27.577267786022276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.007955819368362427, "target_z": -0.5195312459985935, "ce": 6.1894097328186035, "genuineness": 3.9676074981689453, "identity": 0.2381831705570221, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/adddce7e5df726b1e22e213e1e69c4b55cc70a97e43413a7d77cebb4abec5632.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "adfc6ddf5985db779cd9de0ea3988976a5c8283bdcaa4d2581f80ebede0a0421", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "1ba43af58d726ac372ee2f8ab1604eff5f9d96563281dda2cf6355d1469123ed", "generation_seconds_batch": 26.994403230957687, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07546205818653107, "target_z": -0.3906250053352087, "ce": 6.427689552307129, "genuineness": 4.055472373962402, "identity": 0.5760433673858643, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/adfc6ddf5985db779cd9de0ea3988976a5c8283bdcaa4d2581f80ebede0a0421.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "ae1e6dfcc213e4065515957ce1cbaf39e4019b079c4ea5fde946011d1a59e428", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "96415b2f8debb7bff41ac2ba771e62b1e9e18c88c483bb2d760a1730620a8811", "generation_seconds_batch": 29.03429691796191, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.041708946228027344, "target_z": -0.45507811143967775, "ce": 6.181467056274414, "genuineness": 4.137060642242432, "identity": 0.30119088292121887, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ae1e6dfcc213e4065515957ce1cbaf39e4019b079c4ea5fde946011d1a59e428.mp3" }, { "id": "heldout-en-2", "language": "en", "sentences": [ "The road turns left behind the bridge.", "Keep going... until you reach the small wooden gate.", "I will meet you on the other side." ], "text": "The road turns left behind the bridge. Keep going... until you reach the small wooden gate. I will meet you on the other side.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-2", "key": "ae3130a5db36c2790a31e23135fde257111c56fa9c5b8061c1f3e51a8cb61258", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The road turns left behind the bridge.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The road turns left behind the bridge.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Intense Malevolence Malice. 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": "5c54f3dae72510ff6b30f8f8afc0392244b7b923bebc2f3bb8d4e2d46a20bbc5", "generation_seconds_batch": 17.580194295034744, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1327400803565979, "target_z": -0.2812500106704175, "ce": 6.41180419921875, "genuineness": 4.15588903427124, "identity": 0.3703344464302063, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ae3130a5db36c2790a31e23135fde257111c56fa9c5b8061c1f3e51a8cb61258.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": "ae337a46f254325e52bb6ea36179ec00cff3ca5e4f566fb26a62644b61e46d4a", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a701883d16571a96c2aee8188f03dc7950f6ed034770581679cba5f0f29e0b3c", "generation_seconds_batch": 17.308502736967057, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04477740824222565, "target_z": -0.44921874333098905, "ce": 6.0146708488464355, "genuineness": 4.15588903427124, "identity": 0.6766765713691711, "hypothesis": "De weg gaat linksaf naar de brug. Loop verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ae337a46f254325e52bb6ea36179ec00cff3ca5e4f566fb26a62644b61e46d4a.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "aeb0c9ecf8098a4dc278af3046c5b49cb88d030027cbd8cbe69f40150f50f9af", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "eeafb5ef840dd4e7ab9a10b5be1b1facca39c28ff04ad4862cebb24fc75ffc2b", "generation_seconds_batch": 26.94950179895386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03250354528427124, "target_z": -0.47265624422019054, "ce": 6.165581703186035, "genuineness": 3.8546390533447266, "identity": 0.3260256052017212, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/aeb0c9ecf8098a4dc278af3046c5b49cb88d030027cbd8cbe69f40150f50f9af.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "aed18da2e4288901ca846ee051c8e5da704c6cfc26490026580dd191c7b6e36a", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "b69b9ea491740574674b55be1f7504f164c889d661433318c39bef62ef86c4ad", "generation_seconds_batch": 26.90118142508436, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.052959978580474854, "target_z": -0.43359375222300367, "ce": 6.443574905395508, "genuineness": 4.080576419830322, "identity": 0.5698114633560181, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/aed18da2e4288901ca846ee051c8e5da704c6cfc26490026580dd191c7b6e36a.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "af5fe2a506171018ce3411a1d4f2c2d0ec4469df11bcc66d743f72a5e258fcd0", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9d51b203a09e6e4dd797afd19604f85b7f9d1a1b752bd1251b5d4616b1ed9580", "generation_seconds_batch": 23.317241381970234, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06421102583408356, "target_z": -0.41210936455188285, "ce": 6.046441555023193, "genuineness": 4.231201171875, "identity": 0.03954162448644638, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/af5fe2a506171018ce3411a1d4f2c2d0ec4469df11bcc66d743f72a5e258fcd0.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "afa714ce1100becc7965bbb73e128e6663d12b70603405d9e7a7437fb46ecc38", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "a857f17572a233d7bc523e0c50828aa38c733a467eb1647ac22ce1494444ceb1", "generation_seconds_batch": 28.450004700105637, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03864046931266785, "target_z": -0.46093750800281313, "ce": 6.070269584655762, "genuineness": 4.325342178344727, "identity": 0.5097265243530273, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/afa714ce1100becc7965bbb73e128e6663d12b70603405d9e7a7437fb46ecc38.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": "afeb76ec51eacd6443641f83f82c6dc53666495b8db9f8ce9b636cf22d17d1d8", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6140830790024471f2a8a7b4b9f8273c7e32a35f3edd874142d9a619b4b6283e", "generation_seconds_batch": 17.12099462095648, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05807408690452576, "target_z": -0.42382812922370694, "ce": 6.165581703186035, "genuineness": 4.256305694580078, "identity": 0.5410963296890259, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi, quand nous pourrons parler ?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/afeb76ec51eacd6443641f83f82c6dc53666495b8db9f8ce9b636cf22d17d1d8.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "b133726bd51cb1525116fddce761d63d50fe2ab355145a9c6f84144eb1b7fec5", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "4fa4e8c450dfaddde6a324c34f213a7a3757d9688b0d791efb40d2946687f74f", "generation_seconds_batch": 26.965452993987128, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02022966742515564, "target_z": -0.49609377356383866, "ce": 5.935244083404541, "genuineness": 3.9425034523010254, "identity": 0.37742874026298523, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b133726bd51cb1525116fddce761d63d50fe2ab355145a9c6f84144eb1b7fec5.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "b153f6e21154062d530cff6dbddcfff518b2640b3cffe013b8885cf9dddbdcd6", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "60696baf42fd518455014bb7c7a0d1c96b7ef4dd5dffd22891163182fc478afd", "generation_seconds_batch": 26.883039213018492, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01045498251914978, "target_z": -0.5546875115596189, "ce": 6.23706579208374, "genuineness": 4.632867813110352, "identity": 0.36668869853019714, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b153f6e21154062d530cff6dbddcfff518b2640b3cffe013b8885cf9dddbdcd6.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "b2537d370a97bdb957b3867b5cdc3a8c37e0d5d3d0c7fe2fdb2a5421128c9db8", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "93e5d2f9781050e69b52935cdc8840f34df7470a3c180cd8de9baea1573e253c", "generation_seconds_batch": 27.24132507899776, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10205543041229248, "target_z": -0.3398437486661978, "ce": 6.348263263702393, "genuineness": 3.9425034523010254, "identity": 0.5612314939498901, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b2537d370a97bdb957b3867b5cdc3a8c37e0d5d3d0c7fe2fdb2a5421128c9db8.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "b2eff95f41ef8e83d45344e608170d4d0b4c4964309273f9f3550d03b2374f3b", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d071fb8d435c2ab624cf49121b5c1e0062ef96c3e675a9f6ec97064fe0d64386", "generation_seconds_batch": 20.342931573977694, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.024774491786956787, "target_z": -0.5820312673394284, "ce": 5.95112943649292, "genuineness": 4.293961524963379, "identity": 0.8004187345504761, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b2eff95f41ef8e83d45344e608170d4d0b4c4964309273f9f3550d03b2374f3b.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "b31020694645a432549ece8a2563d74b0ead7dcc1968e85a074d07609ef2781d", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "5536479403d94dc12ed16f527fd44109c4882e94a2089daa190c08b6418ceadb", "generation_seconds_batch": 22.902778981020674, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00636368989944458, "target_z": -0.5468750017784029, "ce": 6.340320587158203, "genuineness": 4.413206577301025, "identity": 0.6870045065879822, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b31020694645a432549ece8a2563d74b0ead7dcc1968e85a074d07609ef2781d.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": "b32a33521f0a4142b39e07db298be5784415540fe32af9fcc7101e54ebf8fbc9", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8aed6fb274571d7fcbc4a5ca36fa048a431fff8f46a18fc05ee2221d7cd3fda1", "generation_seconds_batch": 23.347755653085187, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.0329570472240448, "target_z": -0.5976562299929672, "ce": 6.205295085906982, "genuineness": 4.268857479095459, "identity": 0.6573383212089539, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna, naast het raam, neem ze mee als je weggaat.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b32a33521f0a4142b39e07db298be5784415540fe32af9fcc7101e54ebf8fbc9.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": "b37b70a350e04eeb59235bc117eb69f5d4fa4fc390e5c0ab6097e1d7cc5868db", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6332626e305edec0feba73311bc4b9232a17ad35f4ef846b1a705acf940eb549", "generation_seconds_batch": 22.108712989953347, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.05341348052024841, "target_z": -0.6367187219901541, "ce": 6.229123115539551, "genuineness": 4.356721878051758, "identity": 0.6272375583648682, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai regardé dans le tiroir, puis près de la fenêtre. Prends les quand tu partiras.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b37b70a350e04eeb59235bc117eb69f5d4fa4fc390e5c0ab6097e1d7cc5868db.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "b37c9db4025a2e30aab03d6e917a7435974b18b7c186b217931c316ad7beeb74", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "6c525c18fea8752d5eee3f5cb8a2ad7880ada77937289f3b49bc7af28c9e87eb", "generation_seconds_batch": 26.666601025965065, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01045498251914978, "target_z": -0.5546875115596189, "ce": 6.300607204437256, "genuineness": 4.369274139404297, "identity": 0.49640101194381714, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b37c9db4025a2e30aab03d6e917a7435974b18b7c186b217931c316ad7beeb74.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "b3c3fed2a1c0865be127abc7259152a4424ff177c743d9ebd8162b4a54415f56", "bucket": "emo_Malevolence_Malice", "phase": "wording", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "fe7c11c8c207515983031fc0d119f9bfcaf1738d170015b8fcb63d02668b7ea4", "generation_seconds_batch": 20.373846941976808, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.010001450777053833, "target_z": -0.5156250195624321, "ce": 6.213237762451172, "genuineness": 4.243753433227539, "identity": 0.7395853400230408, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b3c3fed2a1c0865be127abc7259152a4424ff177c743d9ebd8162b4a54415f56.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "b3c5a8d20d17978c3de650263fca64d6befd8579a2008079e351e4d0010e30dc", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "bd8ed2fbe5392b984041520cdc2881d4995ddc21318a0ded7791aa5cfe1c919e", "generation_seconds_batch": 19.56057229801081, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.037617653608322144, "target_z": -0.4628906212208938, "ce": 6.070269584655762, "genuineness": 4.306513786315918, "identity": 0.4686674475669861, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b3c5a8d20d17978c3de650263fca64d6befd8579a2008079e351e4d0010e30dc.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": "b3e762068cc8bbce90324a4007bfe61b64d9eaad674327ba985a7733ac715a67", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "034041c421a5965e89be1fb4768134ffed08b5d0945c62a74493fd82f99c50ba", "generation_seconds_batch": 17.211229598033242, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04682305455207825, "target_z": -0.44531248844038107, "ce": 6.070269584655762, "genuineness": 4.231201171875, "identity": 0.7735013365745544, "hypothesis": "La route tourne à gauche, après le pont, continue, jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b3e762068cc8bbce90324a4007bfe61b64d9eaad674327ba985a7733ac715a67.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "b425906ff8504274ac5ef420a2499e9861751a7af19a3a726038d27a69247617", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "3d4d0227c19411ac9fbebcd18a87210fa1c8884ab17ceee80e1e7d7dc377f03e", "generation_seconds_batch": 25.114715668954886, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.012500613927841187, "target_z": -0.5585937379957803, "ce": 6.38003396987915, "genuineness": 4.193545341491699, "identity": 0.5163668990135193, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b425906ff8504274ac5ef420a2499e9861751a7af19a3a726038d27a69247617.mp3" }, { "id": "heldout-de-1", "language": "de", "sentences": [ "Die Schlüssel liegen auf dem Küchentisch.", "Ich suchte in der Schublade... dann neben dem Fenster.", "Nimm sie mit, wenn du gehst." ], "text": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade... dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-1", "key": "b429ad869b4f073af736e4e6510807542fc4f5324b0eb7f096c719d9002f4f71", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Intense Malevolence Malice. 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": "5d68bec0c7497b654bb0f6f877d147d4b810c59e794566d7143c49923cf63c6d", "generation_seconds_batch": 18.91091773903463, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.012047111988067627, "target_z": -0.5117187362173774, "ce": 6.372091293334961, "genuineness": 4.381826400756836, "identity": 0.6137940287590027, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade, dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b429ad869b4f073af736e4e6510807542fc4f5324b0eb7f096c719d9002f4f71.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "b52d960fec4cb562187461213b83536759fcce7f391a682a79aba43347a610fc", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "2d063cee40d0b6de541a76250712ad4336fd4a8037a4721d497d6eb237f7f3a3", "generation_seconds_batch": 25.97454882506281, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.07182428240776062, "target_z": -0.6718749875511796, "ce": 6.340320587158203, "genuineness": 4.21237325668335, "identity": 0.3197709321975708, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b52d960fec4cb562187461213b83536759fcce7f391a682a79aba43347a610fc.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "b53778c61ed59655d58f3e31b5718f4426e9d8a9a82f9d4e2578d2d838b254b0", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "655f5e69671d0964fcad28b8b0ae506d4f99e33812cb9393047ff6efe2daac15", "generation_seconds_batch": 25.749805856030434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8773543834686279, "target_z": 1.1406249484263153, "ce": 6.109982967376709, "genuineness": 3.9173994064331055, "identity": 0.1676025539636612, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b53778c61ed59655d58f3e31b5718f4426e9d8a9a82f9d4e2578d2d838b254b0.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "b54171ff4afeca62529c4d1783300b6df3f8ca892848d4f0447338fe5a5a65c7", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "ea86b0dcb5047a54e408a9953c8a00e051bc25d68a91143e6a49fe6aa7c1b74b", "generation_seconds_batch": 28.573859137948602, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.030911415815353394, "target_z": -0.5937500035568058, "ce": 6.109982967376709, "genuineness": 4.7332844734191895, "identity": 0.5037561058998108, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b54171ff4afeca62529c4d1783300b6df3f8ca892848d4f0447338fe5a5a65c7.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "b548992ba6156450795b615cb12867d79835d56b949a44c6bcc7ed5ccea9d9a9", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "6896df6bafbc7cf7877c2ff307a726bd4d4ac6acc68e50b78e002ae8ba987fb4", "generation_seconds_batch": 23.631674370961264, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5807360410690308, "target_z": 0.5742187006493191, "ce": 5.879645347595215, "genuineness": 4.18726921081543, "identity": 0.31532788276672363, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b548992ba6156450795b615cb12867d79835d56b949a44c6bcc7ed5ccea9d9a9.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": "b56d1a89dbecdf6a89e27c001ca51a6cc43858b54de146e896cc66c43042bd53", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5e4ee9095c7d615b3ca5f5c21ddd7559f94de48742988c9063cfe8e273953dcb", "generation_seconds_batch": 17.16027115401812, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.25241023302078247, "target_z": -0.052734398341538265, "ce": 6.324435234069824, "genuineness": 4.268857479095459, "identity": 0.48810163140296936, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade, dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b56d1a89dbecdf6a89e27c001ca51a6cc43858b54de146e896cc66c43042bd53.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": "b60d0bceb9d2934c1e49e7ea0de61531ab11f51ff2acd9fc3186f9e7bf67b462", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c52c5636d809f1f6c12378294a0988035df3bfb3f2b360b7c621daa0195999a2", "generation_seconds_batch": 18.956564913038164, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7528899908065796, "target_z": 2.812500106704175, "ce": 6.308549880981445, "genuineness": 4.344170093536377, "identity": 0.6732803583145142, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal. Dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b60d0bceb9d2934c1e49e7ea0de61531ab11f51ff2acd9fc3186f9e7bf67b462.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "b6d0b651ecc9a9626094a52cffd78f36897024034493067edb967754508b1467", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "4b72f820d58efd9dc63b1212c494b67588021a52808ab4f633b96a655692e9b4", "generation_seconds_batch": 26.350344324018806, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5155953168869019, "target_z": 2.359375108482578, "ce": 6.205295085906982, "genuineness": 4.243753433227539, "identity": 0.5847368240356445, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b6d0b651ecc9a9626094a52cffd78f36897024034493067edb967754508b1467.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": "b6efe0f2947a96c9a0a1ae7f92a6b88d00f3db4573e140167a04aaae40193c09", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6ed0cd3771bf6bda4943e123b6630f040542a529997f1046b422673b38f0699a", "generation_seconds_batch": 19.05137082596775, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.022275328636169434, "target_z": -0.49218749021878394, "ce": 6.268836498260498, "genuineness": 4.293961524963379, "identity": 0.5708619952201843, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí, junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b6efe0f2947a96c9a0a1ae7f92a6b88d00f3db4573e140167a04aaae40193c09.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "b71d623efc6f603f37a42fd80bb596c17f6886a0e1cc8e2290f153d7edec077f", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "ba88e173a6199fc1bd8632af9090f3882991e71bee0f9795e1abd0aba29cc331", "generation_seconds_batch": 24.149918518029153, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01454627513885498, "target_z": -0.562500021340835, "ce": 6.133810997009277, "genuineness": 4.532451152801514, "identity": 0.7663341760635376, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b71d623efc6f603f37a42fd80bb596c17f6886a0e1cc8e2290f153d7edec077f.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "b73a37b06fdb660283ec52a0e66e7e8f89fe982ddfdb2e02574832c21c82f969", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "289618d10125823ff8cae0b47444be1bdea2e0075b367d9b2299e56331d17d43", "generation_seconds_batch": 28.571292817941867, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00022676587104797363, "target_z": -0.5351562655610255, "ce": 6.141753673553467, "genuineness": 4.2751336097717285, "identity": 0.643538236618042, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b73a37b06fdb660283ec52a0e66e7e8f89fe982ddfdb2e02574832c21c82f969.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "b7b318ca51bb7b4a47e80d7096dc31afa9af6e6a610691c0bffd1e081f0dc214", "bucket": "emo_Malevolence_Malice", "phase": "wording", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "b419dbf6b5872cafac9ce1371eb35ec53ab49f747c7685dcf1f435e776c29924", "generation_seconds_batch": 21.299496843013912, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3424185812473297, "target_z": 0.11914064611853463, "ce": 6.300607204437256, "genuineness": 4.1998209953308105, "identity": 0.42184382677078247, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b7b318ca51bb7b4a47e80d7096dc31afa9af6e6a610691c0bffd1e081f0dc214.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "b835f43e6ffd36f294de4731a991216ac8ccde4bd3a7558168d47463099cf718", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "a6f01513d5022e638ee6f579e392cfb42d70d1340a57ee74c01c5f886fb2d193", "generation_seconds_batch": 27.838630085927434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.081918716430664, "target_z": 1.5312498683981843, "ce": 6.157639026641846, "genuineness": 4.18099308013916, "identity": 0.5382490158081055, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b835f43e6ffd36f294de4731a991216ac8ccde4bd3a7558168d47463099cf718.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "b8483e00b08cff2535b43002c7159f90facf662cf18f599987e76ebe8f08d8b9", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "7a1bfcab252eb99cd401f9ea90dfe9f9956cb49b4ea48b8ae31c617a57c6cd81", "generation_seconds_batch": 19.42116375407204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2242826521396637, "target_z": -0.10644529638322357, "ce": 6.141753673553467, "genuineness": 4.570107460021973, "identity": 0.5742862224578857, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b8483e00b08cff2535b43002c7159f90facf662cf18f599987e76ebe8f08d8b9.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "b85744a8d016a828aae7843924164222cca0b3ecc630d979ae495d147a68fedf", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0e1a48d7a8327ec0dfcb151c308955249b92bf15b9f706c33a919e615cc20565", "generation_seconds_batch": 26.80087589903269, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03454919159412384, "target_z": -0.4687499893295825, "ce": 6.24500846862793, "genuineness": 4.024092197418213, "identity": 0.5208510160446167, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b85744a8d016a828aae7843924164222cca0b3ecc630d979ae495d147a68fedf.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "b884076f4a37ab9ea2679155bf507eb6ef38996237a69969e61f1c44410b0dea", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9a6bf90b7f9342bd68a338ff0782719f417c141f65a8050967ea3143b5d27541", "generation_seconds_batch": 27.479947610991076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.052959978580474854, "target_z": -0.43359375222300367, "ce": 5.685050010681152, "genuineness": 3.7918784618377686, "identity": 0.5696828365325928, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b884076f4a37ab9ea2679155bf507eb6ef38996237a69969e61f1c44410b0dea.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "b8a7c5301c35026d63af55a20cc2df10e1748456ec6affd83b6c05bf627c86ac", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "8f01038b0b065a83a2fb206da5156098b52d6628ab2869a7a8701cd3c1523304", "generation_seconds_batch": 28.61614420299884, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.018184036016464233, "target_z": -0.5, "ce": 6.086154937744141, "genuineness": 4.444586753845215, "identity": 0.5488262176513672, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b8a7c5301c35026d63af55a20cc2df10e1748456ec6affd83b6c05bf627c86ac.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "b8aee6aad7015299ce7f09271fccc7c1f4ee17825ad7a09bbfa545d0bc7cc91c", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "84bdd886745b7df2d3c42a78ebbd78e01378395cc0d5339a9ee9d983faa89449", "generation_seconds_batch": 27.55493992008269, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.010001450777053833, "target_z": -0.5156250195624321, "ce": 6.173524379730225, "genuineness": 4.545003414154053, "identity": 0.5704283714294434, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b8aee6aad7015299ce7f09271fccc7c1f4ee17825ad7a09bbfa545d0bc7cc91c.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "b95805674c93a3e8c3a82527b7437dc77983ea3e8fe45eed97af2fe90a02c09e", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "19d609764702d12c0005afd13e0dfd5717c87d9037af670071362188663fa07f", "generation_seconds_batch": 27.352845833054744, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.018184036016464233, "target_z": -0.5, "ce": 6.109982967376709, "genuineness": 4.268857479095459, "identity": 0.657593846321106, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b95805674c93a3e8c3a82527b7437dc77983ea3e8fe45eed97af2fe90a02c09e.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "b961759998b898bd2e00ebce18b8270ef868286ecd58acc65db534c08ee2961c", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "d45ecc9db485f9e597817b6e4564c69586144eed1e4f9a5e5d8c7eef2cbf8531", "generation_seconds_batch": 18.91052573500201, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.156264990568161, "target_z": -0.23632812211009527, "ce": 5.935244083404541, "genuineness": 4.093128204345703, "identity": 0.425915002822876, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b961759998b898bd2e00ebce18b8270ef868286ecd58acc65db534c08ee2961c.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "b9a7e50e69562dda3252cad8d4e5241ee9267a08bec9693727691098b628e095", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "f33023e3b522bcfd486fba7f4ce6a5bcd969853858fc5cddc83be8eb8d58cdbe", "generation_seconds_batch": 25.513310719979927, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.022275328636169434, "target_z": -0.49218749021878394, "ce": 6.046441555023193, "genuineness": 4.306513786315918, "identity": 0.6428835391998291, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b9a7e50e69562dda3252cad8d4e5241ee9267a08bec9693727691098b628e095.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": "b9c3e0012c624f83eaeb26daa0aa3fa36510fd16206e536d4fc12fecb95f3b95", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d60bf3af0d4a771eedf9087513d5b3e75babc8314017abbc163b5287944b5162", "generation_seconds_batch": 17.239664223976433, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03454919159412384, "target_z": -0.4687499893295825, "ce": 6.268836498260498, "genuineness": 4.256305694580078, "identity": 0.5209733247756958, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/b9c3e0012c624f83eaeb26daa0aa3fa36510fd16206e536d4fc12fecb95f3b95.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "ba1031b92a3452f76d391010c0f7b5f637422c737671e3cb82d293b550ee6cf1", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "1654e36ecb2f1a90c547a6591f89429e1e10ab14375f7187797b5b070110465c", "generation_seconds_batch": 26.71569083002396, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.016591906547546387, "target_z": -0.5664062477769963, "ce": 6.181467056274414, "genuineness": 4.262581825256348, "identity": 0.5287440419197083, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ba1031b92a3452f76d391010c0f7b5f637422c737671e3cb82d293b550ee6cf1.mp3" }, { "id": "heldout-fr-1", "language": "fr", "sentences": [ "Les clés sont sur la table de la cuisine.", "J’ai regardé dans le tiroir... puis près de la fenêtre.", "Prends-les quand tu partiras." ], "text": "Les clés sont sur la table de la cuisine. J’ai regardé dans le tiroir... puis près de la fenêtre. Prends-les quand tu partiras.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-1", "key": "ba70e7e208a6b15a9236ccd5a69e0e5fd403df2a937202d690effa2cdc45d9f2", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Intense Malevolence Malice. 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": "9968b2886a7bab768cfae41d1964b2716399b00bda5cb732772da3b013d82ab2", "generation_seconds_batch": 20.0645887440769, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.0329570472240448, "target_z": -0.5976562299929672, "ce": 6.284721851348877, "genuineness": 4.582659721374512, "identity": 0.5256169438362122, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai regardé dans le tiroir, puis près de la fenêtre. Prends-les quand tu partiras.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ba70e7e208a6b15a9236ccd5a69e0e5fd403df2a937202d690effa2cdc45d9f2.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": "bb56b0764bd704b979455817e646a659ead30679359439c35512456dddc528ce", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The road turns left behind the bridge.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f81c1846b59137c66df3fa9c0bced20f052d6283880f9b27c359bfa8aec92637", "generation_seconds_batch": 20.78920121199917, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.008409321308135986, "target_z": -0.5507812282145643, "ce": 6.419747352600098, "genuineness": 3.923675537109375, "identity": 0.33080774545669556, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/bb56b0764bd704b979455817e646a659ead30679359439c35512456dddc528ce.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "bbc57bba372f0c490ff98203cecc7865f3d16ced379a53227fd365b06f74955b", "bucket": "emo_Malevolence_Malice", "phase": "wording", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "5a700d117b3d2eda02e31be1f089138a830d0bade5485441ccc77b336b6367c4", "generation_seconds_batch": 19.26459373498801, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.026820123195648193, "target_z": -0.5859374937755898, "ce": 6.117925643920898, "genuineness": 4.545003414154053, "identity": 0.48936066031455994, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/bbc57bba372f0c490ff98203cecc7865f3d16ced379a53227fd365b06f74955b.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "bbdd8779da2926fa0dc9be5089ce55c91fc29b8fbc5a5fc8c2d0fee7a3739a2e", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "6eb3eeeefaeae001ae6611f97f203539725be18c808d84ce46622ca031b51bfd", "generation_seconds_batch": 27.59419017098844, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0059101879596710205, "target_z": -0.5234374724347548, "ce": 6.070269584655762, "genuineness": 4.356721878051758, "identity": 0.4674046039581299, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/bbdd8779da2926fa0dc9be5089ce55c91fc29b8fbc5a5fc8c2d0fee7a3739a2e.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "bbfe7910c048aa0598f2b1f8a8f769620e0d4e4a58b880aec57404199e663cca", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "ca9825df662fd0324a05105551c86ba42cbf8cd2774ff9d2b437e264e53f747f", "generation_seconds_batch": 25.10108997800853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1347857266664505, "target_z": -0.27734375577980946, "ce": 6.0146708488464355, "genuineness": 3.9927117824554443, "identity": 0.5497543811798096, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/bbfe7910c048aa0598f2b1f8a8f769620e0d4e4a58b880aec57404199e663cca.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "bc3a7de320e5fe9564f13cc2c4908c79c39eb32bd3f9c8d934413526d3127f99", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "5998f2ce3e6e64d474be46028dd3f0e95303a3a46a329d092a6a64886f4be1a3", "generation_seconds_batch": 25.093174056033604, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.043185263872146606, "target_z": -0.6171874759915607, "ce": 6.260893821716309, "genuineness": 4.444586753845215, "identity": 0.8115571737289429, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/bc3a7de320e5fe9564f13cc2c4908c79c39eb32bd3f9c8d934413526d3127f99.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "bc6c80f719706a5f2a11c4efbea6c9ff0b4b0434dd05c7c5b42bef693f321d31", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "e7cef7ced15aff1db01c268b416efcf921b29fc4bc20311d5dfdd0597b39de19", "generation_seconds_batch": 26.164028721046634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.05750477313995361, "target_z": -0.6445312317713701, "ce": 6.403861999511719, "genuineness": 4.337893962860107, "identity": 0.4828605055809021, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/bc6c80f719706a5f2a11c4efbea6c9ff0b4b0434dd05c7c5b42bef693f321d31.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "bca96a6ddc0386df04fc655e3064a369e5bee35c1a5050a56d1cbb32a387615c", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "237f384271c287d26a9ffc53a6c606e0069ee9f7ca286dc9c9540534ee725b14", "generation_seconds_batch": 28.219675490050577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.016591906547546387, "target_z": -0.5664062477769963, "ce": 6.006728172302246, "genuineness": 4.074300289154053, "identity": 0.6033726930618286, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/bca96a6ddc0386df04fc655e3064a369e5bee35c1a5050a56d1cbb32a387615c.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "bcb4008d0d496455bfb5cf66b6782cf9a97f093b604d6f1e0d9a4bf468e556c6", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "f8f6318837774b0300d3241fdaad38badb6f6b293ba458ce4f1c13473ba07f41", "generation_seconds_batch": 29.309304931084625, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3577609062194824, "target_z": 0.14843751511642478, "ce": 5.990842819213867, "genuineness": 4.049196243286133, "identity": 0.46457165479660034, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tout y est prêt.", "wer": 0.16666666666666666, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/bcb4008d0d496455bfb5cf66b6782cf9a97f093b604d6f1e0d9a4bf468e556c6.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "bceca0dcc69ed7e3dea21c880083c7567f50926c43c824f8a577d791da5a6bc4", "bucket": "emo_Malevolence_Malice", "phase": "wording", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "85e0135f4008589407e7e0ebd9e70b886883ea5fdbed870203458e1d5dcd3e0d", "generation_seconds_batch": 19.298099645064212, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02841225266456604, "target_z": -0.48046875400140654, "ce": 6.348263263702393, "genuineness": 4.482243061065674, "identity": 0.6523622274398804, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/bceca0dcc69ed7e3dea21c880083c7567f50926c43c824f8a577d791da5a6bc4.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "bd692e59bb55ab3809439a5618ebc511f64f467678671e420ec3c4723a08cbf1", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "d1f32fdb1fd2570624499aae5437d5e1cafae7e33db52d339528d0beeafca7c3", "generation_seconds_batch": 18.38138328201603, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08671310544013977, "target_z": -0.36914061766408796, "ce": 6.395919322967529, "genuineness": 4.15588903427124, "identity": 0.5864812135696411, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/bd692e59bb55ab3809439a5618ebc511f64f467678671e420ec3c4723a08cbf1.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": "bd6d47f49dfe5bc48b5607494af6fa8553bbf41f3e5d47e346cdb8393bda9cc2", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "587fb3ca97ce6701b5c774f48f6f0992cf9b222def4b759c959f97c5048a311b", "generation_seconds_batch": 23.243263807031326, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.016138404607772827, "target_z": -0.5039062264361613, "ce": 6.2767791748046875, "genuineness": 4.319066047668457, "identity": 0.5065429210662842, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/bd6d47f49dfe5bc48b5607494af6fa8553bbf41f3e5d47e346cdb8393bda9cc2.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "bea486b55d92361a063a9e81629b6c0a036eab24d6cb9e536813fc4edfa827d4", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "5c13457e3d4885a52507f352f1f54d72a8084288d130141f4846100481f9250e", "generation_seconds_batch": 25.97657679603435, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.055005624890327454, "target_z": -0.42968749733239564, "ce": 6.2767791748046875, "genuineness": 4.595211505889893, "identity": 0.3200124204158783, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/bea486b55d92361a063a9e81629b6c0a036eab24d6cb9e536813fc4edfa827d4.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "beed1bbb8eddd6a50f8ce49b98fc189b7ce9a61bf7a081bf8460bf8df176b1e5", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "08cb5b2b64f9314507d0e34ba8a198c16a51b3f624033c321cc27af26ff68484", "generation_seconds_batch": 21.740140330046415, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00022676587104797363, "target_z": -0.5351562655610255, "ce": 6.300607204437256, "genuineness": 4.381826400756836, "identity": 0.35669663548469543, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/beed1bbb8eddd6a50f8ce49b98fc189b7ce9a61bf7a081bf8460bf8df176b1e5.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "bf5fcd51f33f62b021be2632b375a3eceabafc375f6d62603a5b0507e0a87532", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "6bb4e070e4767053608c9da950a1e027261b17a7908c536c3866ae3140bfeeb7", "generation_seconds_batch": 19.419238707982004, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.028865784406661987, "target_z": -0.5898437771206445, "ce": 6.070269584655762, "genuineness": 4.306513786315918, "identity": 0.6174657344818115, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/bf5fcd51f33f62b021be2632b375a3eceabafc375f6d62603a5b0507e0a87532.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": "c0336e35c75b3c523dddffeb67614cce6da232d7aeb99c4124bc2960f45189fa", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6945c6513234143a9d7279f045dcaf0c29b67979e0a8c94d7bb6a1dce4ece3eb", "generation_seconds_batch": 22.95551992196124, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.304829865694046, "target_z": 0.047363280082923086, "ce": 6.252951145172119, "genuineness": 4.419482707977295, "identity": 0.677150309085846, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c0336e35c75b3c523dddffeb67614cce6da232d7aeb99c4124bc2960f45189fa.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "c056748aa78d653393b06071def72bc30935ce5be8a196c1681ca93055cd34df", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3405e0c274d4d816bdac70c13985df8b80b6e65b0fd5d25809e483bfb2d26c56", "generation_seconds_batch": 20.70586210093461, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.720159649848938, "target_z": 2.750000028454447, "ce": 6.332377910614014, "genuineness": 3.936227321624756, "identity": 0.4419179856777191, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c056748aa78d653393b06071def72bc30935ce5be8a196c1681ca93055cd34df.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "c0600c46790e5c050ff4fefaffc676ace57b39477e9e3254b3a2e00db6a6fdf2", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "bd9cc8c9cec5d5927cfda8122ac38fe4a75290ebbb05e0d70911157233cc625e", "generation_seconds_batch": 27.93442015699111, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3078983426094055, "target_z": 0.05322267664605844, "ce": 6.284721851348877, "genuineness": 4.024092197418213, "identity": 0.4924820065498352, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c0600c46790e5c050ff4fefaffc676ace57b39477e9e3254b3a2e00db6a6fdf2.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "c06eff776aedaf4266991da807827cfc4bb5bf55f59a65971ed04c65d8dbb974", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e0fbeb5b90c3650d5f8b52b870a53af496c7b998c8f662f13c01207baa90d09f", "generation_seconds_batch": 24.829320911085233, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04273176193237305, "target_z": -0.4531249982215971, "ce": 6.133810997009277, "genuineness": 4.080576419830322, "identity": 0.36831581592559814, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c06eff776aedaf4266991da807827cfc4bb5bf55f59a65971ed04c65d8dbb974.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "c0c77509ed18bf5ba0f9580a154cc9c01952a6108490961d0819101e08b92130", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "340676ea0fbe94ce5c2041415810f0dfb521c5a7177745119b0a49b0df6191c3", "generation_seconds_batch": 20.851765237981454, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03250354528427124, "target_z": -0.47265624422019054, "ce": 5.990842819213867, "genuineness": 4.400654315948486, "identity": 0.7029691934585571, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c0c77509ed18bf5ba0f9580a154cc9c01952a6108490961d0819101e08b92130.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "c0cc2d44e81963c02a992e7aa0796109134811b79fb544cac3398937411446de", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "70efc2812af51f13f406b704177a41f163bd3266fdc7ad3aa4e7f335edd928ca", "generation_seconds_batch": 25.94337989005726, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.010001450777053833, "target_z": -0.5156250195624321, "ce": 6.181467056274414, "genuineness": 3.948779582977295, "identity": 0.41968512535095215, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c0cc2d44e81963c02a992e7aa0796109134811b79fb544cac3398937411446de.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "c12d1672f6a8d5503e75717f4801e3d587780760a7ee515ac965017064e4047f", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "5dfbfd38a6f2d16e260e586bf54e23b50fbffe70275b42fb8c12140df6d6ea35", "generation_seconds_batch": 26.091396669042297, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.027389436960220337, "target_z": -0.48242186721948727, "ce": 6.070269584655762, "genuineness": 4.143336772918701, "identity": 0.3821563124656677, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c12d1672f6a8d5503e75717f4801e3d587780760a7ee515ac965017064e4047f.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "c14cd0b15acf87768741cd9dcd296c46bb3b2fcfa9e05470dd6ee175000e7b67", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "4561855750df9b84280dbecb57bf5aacc75ad09298a96e4ba9a283090170c6cd", "generation_seconds_batch": 18.864009695942514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.40992480516433716, "target_z": 0.24804685832747267, "ce": 6.3641486167907715, "genuineness": 4.400654315948486, "identity": 0.33744168281555176, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c14cd0b15acf87768741cd9dcd296c46bb3b2fcfa9e05470dd6ee175000e7b67.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "c202350e387f58a0eee67a5aa98c45ae4241ae5afaf39fd398d05c208cd92066", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "5269ced3d7363f03df3394c01a736dde13ef699246f24eb1c1a688c244ce4b1b", "generation_seconds_batch": 18.862858982989565, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2332963943481445, "target_z": 1.820312445758711, "ce": 6.1894097328186035, "genuineness": 4.293961524963379, "identity": 0.6310313940048218, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c202350e387f58a0eee67a5aa98c45ae4241ae5afaf39fd398d05c208cd92066.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "c202fa9ed0e8ce1aae52d583a5dac79e2aa1f9004a127d57c209d2aae41de538", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "07256cb81d5c9bbcf7eaf2f26439edce82504760b259c9bf6437065a332abf02", "generation_seconds_batch": 27.131395080010407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.020683199167251587, "target_z": -0.5742187575582124, "ce": 5.967014789581299, "genuineness": 4.482243061065674, "identity": 0.652290940284729, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c202fa9ed0e8ce1aae52d583a5dac79e2aa1f9004a127d57c209d2aae41de538.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": "c2876431aff7d2799fe5e69ecd92e72c1083d9a4db2e666f039d57a3bd567eac", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fbd9b9fb85f3da9f5f058d53701be6aa50638737fcf618d6da7d6c30f5b40303", "generation_seconds_batch": 18.91142566106282, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1460367739200592, "target_z": -0.2558593681086887, "ce": 6.205295085906982, "genuineness": 4.093128204345703, "identity": 0.47367599606513977, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, lies sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c2876431aff7d2799fe5e69ecd92e72c1083d9a4db2e666f039d57a3bd567eac.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": "c2b8e7fe215ebc577d83f68c1e19b778391912edc511b3a3e98b35d348e245a3", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "73c65162b4aca8b5a1217ea37242bc42559fbcf1e29a284b2b2a3a5be61fcdfc", "generation_seconds_batch": 19.704933063942008, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2150772511959076, "target_z": -0.12402342916373633, "ce": 6.141753673553467, "genuineness": 4.494794845581055, "identity": 0.74968421459198, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis moi quand nous pourrons parler.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c2b8e7fe215ebc577d83f68c1e19b778391912edc511b3a3e98b35d348e245a3.mp3" }, { "id": "heldout-es-2", "language": "es", "sentences": [ "La carretera gira a la izquierda tras el puente.", "Sigue... hasta la pequeña puerta de madera.", "Te esperaré al otro lado." ], "text": "La carretera gira a la izquierda tras el puente. Sigue... hasta la pequeña puerta de madera. Te esperaré al otro lado.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-2", "key": "c3977db0634952a7f9e4083e6135fbbc87199d34c7cb8c3338515d9ae38c18f4", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Intense Malevolence Malice. 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": "747872b99f8e451fa8c82bdb84967ef5fd309d63c3690f2cf40367c5875b288e", "generation_seconds_batch": 20.12568565295078, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04886868596076965, "target_z": -0.4414062620042197, "ce": 6.252951145172119, "genuineness": 4.545003414154053, "identity": 0.6659923791885376, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c3977db0634952a7f9e4083e6135fbbc87199d34c7cb8c3338515d9ae38c18f4.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "c3a5f61f29fa85a314beac22528e2328782aa90bb3b6ffb68cd051f45ebec9c3", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "e7fa4b7a5729ae25758711699d1f84590ea99bfc6cf007306417a5142f94e0b4", "generation_seconds_batch": 25.160550824017264, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08875875174999237, "target_z": -0.36523436277348, "ce": 6.41180419921875, "genuineness": 4.118232727050781, "identity": 0.5038655996322632, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c3a5f61f29fa85a314beac22528e2328782aa90bb3b6ffb68cd051f45ebec9c3.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "c3bfa48a7e26b16e615f7d4143bcbd76afaeebb23e6387af1293f466a4b85c26", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fb0aa68f139e3adcbaf7a644841f9dd139e0a28aedc347fcb6fed28ea9d26318", "generation_seconds_batch": 24.60730907099787, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01454627513885498, "target_z": -0.562500021340835, "ce": 5.824047088623047, "genuineness": 4.256305694580078, "identity": 0.7095170021057129, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c3bfa48a7e26b16e615f7d4143bcbd76afaeebb23e6387af1293f466a4b85c26.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "c42d9abea06891a22a944eb0f0aa05561d1ef54cad65c73d87540e9337a2d079", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "daa896da0391c6adc3dd1aaa0e88b4d0e526e5700e12d38163871b8b28e6ea67", "generation_seconds_batch": 31.24500945303589, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.0493222177028656, "target_z": -0.6289062691178313, "ce": 6.395919322967529, "genuineness": 4.093128204345703, "identity": 0.5411779880523682, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c42d9abea06891a22a944eb0f0aa05561d1ef54cad65c73d87540e9337a2d079.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "c4476cc104d25f6e457b3058afa5e4577b38ec82a9398e79595fec311789d078", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "b54a066eb7aa8938aed87fca46430d5e6965f609c09f732523901c2e5253c01e", "generation_seconds_batch": 22.66036220395472, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.026366621255874634, "target_z": -0.48437498043756794, "ce": 5.847874641418457, "genuineness": 3.8797430992126465, "identity": 0.29088544845581055, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c4476cc104d25f6e457b3058afa5e4577b38ec82a9398e79595fec311789d078.mp3" }, { "id": "heldout-nl-3", "language": "nl", "sentences": [ "Iemand heeft het kleine blauwe stoeltje verplaatst.", "Het stond hier... naast de hoge boekenkast.", "We zetten het na de lunch terug." ], "text": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier... naast de hoge boekenkast. We zetten het na de lunch terug.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-3", "key": "c493619abd8899c9b3ce96f2a00a6bcdd900e874ef21436707b986365a8437ee", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Intense Malevolence Malice. 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": "3feaedb89c7bfdfd8f5ba5563d7306a68bb437f3520f6b73c755f8d2b79bd4c2", "generation_seconds_batch": 19.02442735200748, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02125251293182373, "target_z": -0.49414060343686467, "ce": 6.229123115539551, "genuineness": 4.231201171875, "identity": 0.7203771471977234, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c493619abd8899c9b3ce96f2a00a6bcdd900e874ef21436707b986365a8437ee.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "c49d52ceb19eb5b3aa261137ef783f157f5bf4e772ed31ae8ba453e74adc5fe0", "bucket": "emo_Malevolence_Malice", "phase": "wording", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "916a621c116fa87b184c06a2b163558a626cd4638982f412bd8955204d030912", "generation_seconds_batch": 20.373846941976808, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.025343775749206543, "target_z": -0.48632815056454193, "ce": 6.173524379730225, "genuineness": 4.494794845581055, "identity": 0.5619785785675049, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c49d52ceb19eb5b3aa261137ef783f157f5bf4e772ed31ae8ba453e74adc5fe0.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "c55be669c05f454ba8cb15dfe09b1c4c9e63ba14b2646ef7acc8cd80bffabe91", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3242c775216176cd4695941e28ef48b201399c31cf7af21b2c868462dabf8e66", "generation_seconds_batch": 26.965452993987128, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07443924248218536, "target_z": -0.39257811855328945, "ce": 5.506340026855469, "genuineness": 3.873466968536377, "identity": 0.6749686002731323, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c55be669c05f454ba8cb15dfe09b1c4c9e63ba14b2646ef7acc8cd80bffabe91.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "c57ae0c7fbc41c57d05dd969a0a35a42d44e53edec3764cd3f1b7af059a02a2a", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "4cd59e5f0ae0f5ccece10b099f844b6b71ac798a2ace9711ad587f30083bf260", "generation_seconds_batch": 18.499824290978722, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.037617653608322144, "target_z": -0.4628906212208938, "ce": 6.475345611572266, "genuineness": 4.086852550506592, "identity": 0.5302704572677612, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c57ae0c7fbc41c57d05dd969a0a35a42d44e53edec3764cd3f1b7af059a02a2a.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "c59df899f840c93f65e1b76712f74aed103711dac3477139f6a9f68852592111", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "5d6bcc18ad8b55a2c3659cc8fd9543464358f4a77358ecb99ba7f1aacd0be85a", "generation_seconds_batch": 18.91790888097603, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.0329570472240448, "target_z": -0.5976562299929672, "ce": 6.459460258483887, "genuineness": 4.231201171875, "identity": 0.408329576253891, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c59df899f840c93f65e1b76712f74aed103711dac3477139f6a9f68852592111.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "c6260a6d98a69e66022804d6061b4620e1c163d73a4392aac77f105ff43d5bd8", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "c05d7ee9ccc01003079b4e1ef54a7aacb51cd0b1eb4aa6ae616729758df2c5eb", "generation_seconds_batch": 26.08786106808111, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.06364172697067261, "target_z": -0.6562500248976408, "ce": 6.316492557525635, "genuineness": 4.13078498840332, "identity": 0.28277814388275146, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c6260a6d98a69e66022804d6061b4620e1c163d73a4392aac77f105ff43d5bd8.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "c638d70b4aa5a95d11e5bf6cdcf2701d5c1b08cb35d034712d3266e146b50a0d", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "c713332e58149791d7f34a504731accb6b30cdd96ac1428c5e1be68a2f6f588d", "generation_seconds_batch": 27.70254610606935, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.014092743396759033, "target_z": -0.5078125097812161, "ce": 5.847874641418457, "genuineness": 3.9801597595214844, "identity": 0.46016520261764526, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c638d70b4aa5a95d11e5bf6cdcf2701d5c1b08cb35d034712d3266e146b50a0d.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "c71b26fe039d68bc5edef3bbce151e94cf6137d9bf7b0208d5740629094e47a7", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b3a24ede5d4964b7833982b4b6ad93a434c642c17600aba619970ec64ab91578", "generation_seconds_batch": 25.692224894068204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.014092743396759033, "target_z": -0.5078125097812161, "ce": 6.292664527893066, "genuineness": 4.168440818786621, "identity": 0.4881446659564972, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c71b26fe039d68bc5edef3bbce151e94cf6137d9bf7b0208d5740629094e47a7.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "c736449a2e87c1cab539bad2cd679ac87ddd54df0a9882f379f8918c3ed5a996", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "7d6f3ff19ef9060a828ee660b6e5d0a12895b453adb581ba6d2da20f5467a594", "generation_seconds_batch": 20.70586210093461, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6646074652671814, "target_z": 0.7343750088920146, "ce": 6.24500846862793, "genuineness": 4.086852550506592, "identity": 0.4088178277015686, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c736449a2e87c1cab539bad2cd679ac87ddd54df0a9882f379f8918c3ed5a996.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "c7497a25187fcac477cef999e3078eae160769717e3109d83a5c2da425e81756", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "85c9b8abd80930e3a4ba89cd22686af46f0ecb298ba6da8b0fc4e02ea2e1e67f", "generation_seconds_batch": 29.309304931084625, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.026366621255874634, "target_z": -0.48437498043756794, "ce": 5.784333229064941, "genuineness": 4.055472373962402, "identity": 0.10541732609272003, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c7497a25187fcac477cef999e3078eae160769717e3109d83a5c2da425e81756.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "c7a6baa04b092dbb02293e662cdd13298e136e7504e492ed3e3a479fb2f0ea4d", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "5b521e78c39c0f5efe9db854cb5fa8474101f21d40844c7e7e7e93fba68b0024", "generation_seconds_batch": 26.90118142508436, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1358085572719574, "target_z": -0.27539061410728216, "ce": 6.356205940246582, "genuineness": 4.030367851257324, "identity": 0.4921317398548126, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c7a6baa04b092dbb02293e662cdd13298e136e7504e492ed3e3a479fb2f0ea4d.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": "c7ccfe4cd5c116cf6b7de121b23d1bd7142e355014b972c0a510bb062daa29cd", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7a54c71dc5e1a5eb5419d1eb869cdfa20f6cd4308ce4aeb59b75fae930af67c0", "generation_seconds_batch": 17.2846324989805, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.008409321308135986, "target_z": -0.5507812282145643, "ce": 6.38003396987915, "genuineness": 4.482243061065674, "identity": 0.533582329750061, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c7ccfe4cd5c116cf6b7de121b23d1bd7142e355014b972c0a510bb062daa29cd.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "c7e973e1537d9f1af06a6db21a2904f100684387b7017198e02521e995ac2b5a", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "36299fff5065e7d72e60cba6cbf3394578ec6c27424fad5e525439e9d30bce99", "generation_seconds_batch": 17.27802036900539, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.014092743396759033, "target_z": -0.5078125097812161, "ce": 5.959072113037109, "genuineness": 4.18099308013916, "identity": 0.7340702414512634, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c7e973e1537d9f1af06a6db21a2904f100684387b7017198e02521e995ac2b5a.mp3" }, { "id": "heldout-nl-0", "language": "nl", "sentences": [ "Je bericht kwam nadat ik vertrokken was.", "Ik las het twee keer... en legde mijn telefoon weg.", "Zeg maar wanneer we kunnen praten." ], "text": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer... en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-0", "key": "c82131bfaf89850d394bf3a21117deb94436b65306054c3e1cbac0c65e2f913d", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Intense Malevolence Malice. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "a344ee4c8f6acc2ff72c414b2be59748433c92df6d4ebc39b7d460a016260ab1", "generation_seconds_batch": 22.45649806095753, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02432096004486084, "target_z": -0.4882812637826226, "ce": 6.292664527893066, "genuineness": 4.444586753845215, "identity": 0.7709590196609497, "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/emo_Malevolence_Malice/c82131bfaf89850d394bf3a21117deb94436b65306054c3e1cbac0c65e2f913d.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "c872e96a4c2eb02b1e2c44772869d955336373ee0ae53a311fbb81c328bd9dd1", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "e818840abcd2bf066190727711a4d0adceba02784d4b9fcc89e7a01d45714654", "generation_seconds_batch": 24.82485379301943, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.14961664378643036, "target_z": -0.24902344339095966, "ce": 6.332377910614014, "genuineness": 3.8295347690582275, "identity": 0.4256036877632141, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c872e96a4c2eb02b1e2c44772869d955336373ee0ae53a311fbb81c328bd9dd1.mp3" }, { "id": "heldout-en-3", "language": "en", "sentences": [ "Someone moved the little blue chair.", "It used to stand here... beside the tall bookshelf.", "We can put it back after lunch." ], "text": "Someone moved the little blue chair. It used to stand here... beside the tall bookshelf. We can put it back after lunch.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-3", "key": "c8907269f81c0a33f09ba3b9d9f76b6acb2a8250e6881dbc11786cea01719e9d", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Someone moved the little blue chair.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Someone moved the little blue chair.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Intense Malevolence Malice. 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": "a3e9bf72bd8788c807afaa2f9e2721a4d3559b77ba86ce9d864111bb96740a06", "generation_seconds_batch": 23.75255783495959, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.008409321308135986, "target_z": -0.5507812282145643, "ce": 6.403861999511719, "genuineness": 4.243753433227539, "identity": 0.5430105924606323, "hypothesis": "Someone moved the little blue chair. It used to stand here, beside the tall bookshelf. We can put it back after lunch.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c8907269f81c0a33f09ba3b9d9f76b6acb2a8250e6881dbc11786cea01719e9d.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "c92c4937fc7a43596c9fdcaf9747a483f2ddb7b5f7dabfd1f983285e5da38448", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "df0ef062207a41ac28c051147803f883c60d2eed6ee428fce0d546c84d8e193d", "generation_seconds_batch": 27.59419017098844, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00227239727973938, "target_z": -0.5390624919971869, "ce": 6.086154937744141, "genuineness": 4.557555198669434, "identity": 0.6662375330924988, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c92c4937fc7a43596c9fdcaf9747a483f2ddb7b5f7dabfd1f983285e5da38448.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "c9d88f3ada2ec76c7464b3bd20215adf807cc05f64143acbf464a9970067ef32", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "f43f80f8272763f4d397e46825c8e836454148a98822edcf1eaec5712b2df4f1", "generation_seconds_batch": 29.813897681073286, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.019206851720809937, "target_z": -0.49804688678191933, "ce": 6.141753673553467, "genuineness": 4.356721878051758, "identity": 0.5412678718566895, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/c9d88f3ada2ec76c7464b3bd20215adf807cc05f64143acbf464a9970067ef32.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "ca754e90499e4e8adc5d7faaac05c3cd015a0f1defe7b902aa0d42c20238e2ec", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "8a905a3d5bd122f27f76fc13ff8a5e9b49732de4f85be65ae2d5308012c60a5f", "generation_seconds_batch": 25.8792836680077, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.035002708435058594, "target_z": -0.6015625133380219, "ce": 6.213237762451172, "genuineness": 4.532451152801514, "identity": 0.5492983460426331, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ca754e90499e4e8adc5d7faaac05c3cd015a0f1defe7b902aa0d42c20238e2ec.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "cb8f0331597eaca8b7b53bb8b4fcbe7e74af97531156ce4b97dea3ec60f4df31", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "d77cfd699051370a560f1239f701f6d8d1436fb41e79b611b649a0580948c39a", "generation_seconds_batch": 20.740487104980275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.004318058490753174, "target_z": -0.5429687753422415, "ce": 6.022613525390625, "genuineness": 4.030367851257324, "identity": 0.6910192966461182, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/cb8f0331597eaca8b7b53bb8b4fcbe7e74af97531156ce4b97dea3ec60f4df31.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "cc58e9a1d8396d8a2a8fa31dd7e48330cf0887b43af1aea43b6cc1f1d82478da", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "bc58286b9d0c3fe3824c39b29186d936692e8aac86c002373638e82c99ccecde", "generation_seconds_batch": 29.620181945036165, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.29377061128616333, "target_z": 0.02624512959463909, "ce": 6.181467056274414, "genuineness": 3.8922953605651855, "identity": 0.1515844464302063, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/cc58e9a1d8396d8a2a8fa31dd7e48330cf0887b43af1aea43b6cc1f1d82478da.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "cc5c2651e589a18428d7bb4f472aa43451d11e9670be5d379b885618350da323", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "b77953535f75e13e1ca8170e28c0ca340f5ac2e3e0574b30d3752a27d9cf3c14", "generation_seconds_batch": 29.637260803952813, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04682305455207825, "target_z": -0.44531248844038107, "ce": 6.1894097328186035, "genuineness": 4.545003414154053, "identity": 0.6016651391983032, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/cc5c2651e589a18428d7bb4f472aa43451d11e9670be5d379b885618350da323.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "cc95ea80731b338540eb476b0b5d7103e6e98b04317cd35a58c58eaefcd94c03", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "12051af88ec5f56cd196ee523ee2f95d568dd9d39192763a79eb1bb1a135ffd0", "generation_seconds_batch": 20.636983719072305, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02022966742515564, "target_z": -0.49609377356383866, "ce": 6.221180438995361, "genuineness": 4.218649387359619, "identity": 0.6792376637458801, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/cc95ea80731b338540eb476b0b5d7103e6e98b04317cd35a58c58eaefcd94c03.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "cd9447e4e8c32c2f418725cb65dd28530a75d8924989fc7098884a395ede7113", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "466e897a6f8f1680363875ae87ee51d720c3c57f6943a8de3c556d30e1b7a971", "generation_seconds_batch": 25.800091105978936, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00022676587104797363, "target_z": -0.5351562655610255, "ce": 6.205295085906982, "genuineness": 4.369274139404297, "identity": 0.7517300844192505, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/cd9447e4e8c32c2f418725cb65dd28530a75d8924989fc7098884a395ede7113.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "ce380ae38b6c0bdf35ac57b7323bea8dc9f45bb8f22ae30a5016721ef8ba0e81", "bucket": "emo_Malevolence_Malice", "phase": "wording", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "926c6b77d4532079985d7624434a3693a8edc6aab4bd56f451586491e42ea521", "generation_seconds_batch": 20.101678684004582, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.026820123195648193, "target_z": -0.5859374937755898, "ce": 6.2767791748046875, "genuineness": 4.457139015197754, "identity": 0.7420116066932678, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ce380ae38b6c0bdf35ac57b7323bea8dc9f45bb8f22ae30a5016721ef8ba0e81.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "ce5a43cd3414b139db9e85390c4001a3dec49bed448de7d0cf7c985576b9f3d3", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "78efefe3c94a249af3717257f930fe138ec96b29a9b890a7873c7950e39f15f1", "generation_seconds_batch": 25.175372371915728, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.22965246438980103, "target_z": -0.09619139507940669, "ce": 6.284721851348877, "genuineness": 4.344170093536377, "identity": 0.4236114025115967, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt, sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ce5a43cd3414b139db9e85390c4001a3dec49bed448de7d0cf7c985576b9f3d3.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "cf0b0380f7c2a96456da28fb9411d664da49866338a8e2dba52742102c075eb0", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "e7825c4af353f6a3c3596bb90bf90e64e9ea19b4a7ac7c980be87791f3f2b61b", "generation_seconds_batch": 26.388338941033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08262181282043457, "target_z": -0.376953127445304, "ce": 6.38797664642334, "genuineness": 4.118232727050781, "identity": 0.5660876035690308, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/cf0b0380f7c2a96456da28fb9411d664da49866338a8e2dba52742102c075eb0.mp3" }, { "id": "heldout-de-3", "language": "de", "sentences": [ "Jemand hat den kleinen blauen Stuhl umgestellt.", "Er stand hier... neben dem hohen Bücherregal.", "Wir stellen ihn nach dem Essen zurück." ], "text": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier... neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-3", "key": "cf68e6506caaaad7e7371e72de9d5417fef89211300487ff83acd2484eaab510", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Intense Malevolence Malice. 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": "fdf0a392d33463d641bff4ca5ad417e9bdbf5fade3159cac8481a6eddf52c357", "generation_seconds_batch": 18.544277752982453, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.026366621255874634, "target_z": -0.48437498043756794, "ce": 6.41180419921875, "genuineness": 4.65797233581543, "identity": 0.5715802907943726, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/cf68e6506caaaad7e7371e72de9d5417fef89211300487ff83acd2484eaab510.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "cfd64ac135e493555037522121b1ceffe9bd41d61212ea31a24b8875faf41319", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "2d5913e400109c8b83c53601debb1dcaa1271cabaaf9311df5d282a7f6fe871f", "generation_seconds_batch": 19.0847464739345, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01454627513885498, "target_z": -0.562500021340835, "ce": 5.998785495758057, "genuineness": 3.9048471450805664, "identity": 0.5988672375679016, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/cfd64ac135e493555037522121b1ceffe9bd41d61212ea31a24b8875faf41319.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "cfe3564d1a3e75032a2b39eb580f46d1ddf843a6b2ff9537fa16542bd7ec4799", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "97b073f8f4731dbc1f2c4f14d140b6ded560c24d161ada38f4e28c913f696aa5", "generation_seconds_batch": 27.307952026021667, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.43038123846054077, "target_z": 0.28710935032465956, "ce": 6.125868320465088, "genuineness": 4.193545341491699, "identity": 0.7071648836135864, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/cfe3564d1a3e75032a2b39eb580f46d1ddf843a6b2ff9537fa16542bd7ec4799.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "cff6a56a9989cef495d42ba86e605690cf6bbc850b65db3e3e61ef4af971cd47", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "557dbceec93d4297f25cb9afc6f16dceac41572793a275547edc70d179af75c1", "generation_seconds_batch": 25.70075239497237, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07137078046798706, "target_z": -0.39843748666197815, "ce": 6.332377910614014, "genuineness": 4.243753433227539, "identity": 0.5392932891845703, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/cff6a56a9989cef495d42ba86e605690cf6bbc850b65db3e3e61ef4af971cd47.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "d01ead0c566644d97c8f16a28439fff45a926d9738d5721ce8157cfaf86cfaac", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "433c72028267603023a12c34a00446e310287f1d29df54d5dfa41b164448744c", "generation_seconds_batch": 27.271086927037686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.05341348052024841, "target_z": -0.6367187219901541, "ce": 6.324435234069824, "genuineness": 4.3881025314331055, "identity": 0.6067814826965332, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana, vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d01ead0c566644d97c8f16a28439fff45a926d9738d5721ce8157cfaf86cfaac.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "d029895a6ad9e833af24fa49a01a8a65835af6ebf5a2f165a375bbef3104ea3d", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "108e2cbfae8ce06e49a45653debf4fc10ae22b4a2166992d5311cb9f0319cd94", "generation_seconds_batch": 18.91790888097603, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.0452309250831604, "target_z": -0.6210937593366153, "ce": 6.2767791748046875, "genuineness": 4.118232727050781, "identity": 0.34152108430862427, "hypothesis": "I found the little box. It was beside the old window. We can look inside together,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d029895a6ad9e833af24fa49a01a8a65835af6ebf5a2f165a375bbef3104ea3d.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "d0a3629a3a09bd23edbfb3a9a9a28cba8b044689b30f548522df5274264bba76", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "49773b09b724384c2348b55e200882466395515e11091f1e4c1a93bf290d5098", "generation_seconds_batch": 22.902778981020674, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.024774491786956787, "target_z": -0.5820312673394284, "ce": 6.038498878479004, "genuineness": 4.118232727050781, "identity": 0.5188801884651184, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d0a3629a3a09bd23edbfb3a9a9a28cba8b044689b30f548522df5274264bba76.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "d0b7692053152e991a71c74e9a0f78c90906941b3e7b9f9002208b001a2f7d3d", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "6b9dcfeb4a75e0c6c9182703b3529b31d8a19d7f1659a7ac222bb3b2c73482ca", "generation_seconds_batch": 26.56255738704931, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0737361907958984, "target_z": 1.5156249626535387, "ce": 5.982900142669678, "genuineness": 4.030367851257324, "identity": 0.5227230787277222, "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/emo_Malevolence_Malice/d0b7692053152e991a71c74e9a0f78c90906941b3e7b9f9002208b001a2f7d3d.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "d108c4ea68bba874d94472fdf513951f7f822bafabfacf17b7ec9419e4078210", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "ecdcf219834118a8681a86ee373e569e0991412b2664811003d61e13a0042428", "generation_seconds_batch": 19.42116375407204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.014092743396759033, "target_z": -0.5078125097812161, "ce": 6.086154937744141, "genuineness": 4.344170093536377, "identity": 0.5325556993484497, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d108c4ea68bba874d94472fdf513951f7f822bafabfacf17b7ec9419e4078210.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "d10e520ba438585091e4888ec888eee5d294bc00271757b92e671829eda6a8cb", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "b401e4ad5c46f1f2a2f230cd169c70027108430aef41134c564c67c37a22c42b", "generation_seconds_batch": 27.817798816948198, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03454919159412384, "target_z": -0.4687499893295825, "ce": 6.205295085906982, "genuineness": 4.494794845581055, "identity": 0.5958938598632812, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d10e520ba438585091e4888ec888eee5d294bc00271757b92e671829eda6a8cb.mp3" }, { "id": "heldout-en-4", "language": "en", "sentences": [ "I wrote the number on the envelope.", "Read it carefully... and check the last two digits.", "Call me when you have a moment." ], "text": "I wrote the number on the envelope. Read it carefully... and check the last two digits. Call me when you have a moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-4", "key": "d11caac31fa3732e2517803823f1b039213fb31e9438619fd83d8b232c894b7d", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I wrote the number on the envelope.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I wrote the number on the envelope.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Intense Malevolence Malice. 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": "f47b824291118769c0c5f830a30980acf39abeaaf89d15e3f507f01fb2c05c53", "generation_seconds_batch": 20.750103883910924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02125251293182373, "target_z": -0.49414060343686467, "ce": 6.38797664642334, "genuineness": 3.8295347690582275, "identity": 0.22619211673736572, "hypothesis": "I wrote the number on the envelope. Read it carefully and check the last two digits. Call me when you have a moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d11caac31fa3732e2517803823f1b039213fb31e9438619fd83d8b232c894b7d.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "d1358b633329f3224210032ecece4968b342d73bdca279bfdbdefaca3da065fc", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "04f0eec3a591afddb8cf2090968fc835ad6e6c2e174751dca72e3f4debbd4996", "generation_seconds_batch": 23.3032032570336, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.016138404607772827, "target_z": -0.5039062264361613, "ce": 6.046441555023193, "genuineness": 4.13078498840332, "identity": 0.5769885182380676, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d1358b633329f3224210032ecece4968b342d73bdca279bfdbdefaca3da065fc.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "d172009685548233381540c18cf859008f98454315fa6224e1fe89b384c5910d", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "80ab25344d99e570456910e96340d80c4b127e6c613bc54263c901716644bd94", "generation_seconds_batch": 19.698353365063667, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05602844059467316, "target_z": -0.42773438411431497, "ce": 6.308549880981445, "genuineness": 4.080576419830322, "identity": 0.5940233469009399, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d172009685548233381540c18cf859008f98454315fa6224e1fe89b384c5910d.mp3" }, { "id": "heldout-es-1", "language": "es", "sentences": [ "Las llaves están en la mesa de la cocina.", "Miré en el cajón... y luego junto a la ventana.", "Llévatelas cuando te vayas." ], "text": "Las llaves están en la mesa de la cocina. Miré en el cajón... y luego junto a la ventana. Llévatelas cuando te vayas.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-1", "key": "d2319f0ba4d4a13f099c61fd2524d2cd47bb09bca092815ed8a29e7f5b0c527c", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Intense Malevolence Malice. 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": "686dd5390531b4b67997044d3c51ae743191bd07aef9617ef0e017dc47404d32", "generation_seconds_batch": 22.309804894961417, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01045498251914978, "target_z": -0.5546875115596189, "ce": 6.38003396987915, "genuineness": 4.030367851257324, "identity": 0.6617389917373657, "hypothesis": "Las llaves están en la mesa de la cocina. Miré en el cajón y luego, junto a la ventana, llévatelas cuando te vayas.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d2319f0ba4d4a13f099c61fd2524d2cd47bb09bca092815ed8a29e7f5b0c527c.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "d2bda824408ca75f82692bbed8a00447cff491e68a7da157b78b32c9f710269a", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "96c2d2e17d0727230a59e1bbe36da6e4289869c076d80e9a1302cfaf8d271f8d", "generation_seconds_batch": 25.863272108021192, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.018637537956237793, "target_z": -0.5703124742131577, "ce": 6.181467056274414, "genuineness": 3.9425034523010254, "identity": 0.4166366755962372, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d2bda824408ca75f82692bbed8a00447cff491e68a7da157b78b32c9f710269a.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "d2bf58c0ad1b28c76758674bd9ce24246f35f796d202f1550dc742f29d9bba62", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "afe37aa2ebb0852b9cfed7832f51567c807c1536f4390726f26fae48bdc7aa3c", "generation_seconds_batch": 21.186607319978066, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09489567577838898, "target_z": -0.3535156265561025, "ce": 6.308549880981445, "genuineness": 3.6914620399475098, "identity": 0.35557907819747925, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d2bf58c0ad1b28c76758674bd9ce24246f35f796d202f1550dc742f29d9bba62.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "d2e5ace2d7c13037de9bf07dd231ad1a94198a5f810bb00f9aea6699cc07c5e1", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "a07572cf9a295a1ebab53998a837419023c1d3d9afab04ccc6ae8e83a857f469", "generation_seconds_batch": 25.097128701978363, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.33474740386009216, "target_z": 0.10449218316514289, "ce": 6.0146708488464355, "genuineness": 3.9425034523010254, "identity": 0.4581599533557892, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d2e5ace2d7c13037de9bf07dd231ad1a94198a5f810bb00f9aea6699cc07c5e1.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "d31d1471f9b7f61771e37a58191de90071dc460abd0ee6ea18ec441871fdbb99", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "b8e2456da8cd661c53a6bd65f75b2214cad157d70dda00952b75242a9dc4d232", "generation_seconds_batch": 23.136293446994387, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00227239727973938, "target_z": -0.5390624919971869, "ce": 6.284721851348877, "genuineness": 3.9801597595214844, "identity": 0.32239094376564026, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d31d1471f9b7f61771e37a58191de90071dc460abd0ee6ea18ec441871fdbb99.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "d3308d5569bed04a23b1bf764fd7ec4d312357ffadc3b4e9488736a2cb71e43f", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "a21b31449803a411e4f55c047ad04664b3f1038f593fda4fedabf421d0090472", "generation_seconds_batch": 28.870381204993464, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.18848387897014618, "target_z": -0.17480468583274728, "ce": 6.173524379730225, "genuineness": 4.231201171875, "identity": 0.4773615598678589, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d3308d5569bed04a23b1bf764fd7ec4d312357ffadc3b4e9488736a2cb71e43f.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "d395d2a3c1c90442a04fc31594243ee91fd37e7d7c291ae4a60fd9e231781291", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "2a3ae3c2c35b04ba20d3e2b43bc3d1cdb31584bbf0fd2b91aba4c556c77d0a49", "generation_seconds_batch": 26.350685394951142, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.019206851720809937, "target_z": -0.49804688678191933, "ce": 6.109982967376709, "genuineness": 4.582659721374512, "identity": 0.5911297798156738, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d395d2a3c1c90442a04fc31594243ee91fd37e7d7c291ae4a60fd9e231781291.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "d3f1a07861e851e153e877cdae377fb75f1e249c2366240d726bfde8d54ad642", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "24b8cb3410f99e4ba46e1793ee0d55de49fc8c804ca9fb3f674600fc07789135", "generation_seconds_batch": 27.208889418980107, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0018188953399658203, "target_z": -0.5312499822159709, "ce": 6.284721851348877, "genuineness": 4.607763767242432, "identity": 0.5274955630302429, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d3f1a07861e851e153e877cdae377fb75f1e249c2366240d726bfde8d54ad642.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "d3f8c7421257a7f72bb22adc723c8429c0f58871a31cc087b07c2049c241721e", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "2298bc255abf9d2cedd244ee47a9225b2eebbfa5a4d6f2baefbf832ca3b6865d", "generation_seconds_batch": 29.596475759055465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.024774491786956787, "target_z": -0.5820312673394284, "ce": 5.998785495758057, "genuineness": 4.519899368286133, "identity": 0.7838926911354065, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d3f8c7421257a7f72bb22adc723c8429c0f58871a31cc087b07c2049c241721e.mp3" }, { "id": "heldout-fr-4", "language": "fr", "sentences": [ "J’ai écrit le numéro sur l’enveloppe.", "Lis-le bien... et vérifie les deux derniers chiffres.", "Appelle-moi quand tu auras un moment." ], "text": "J’ai écrit le numéro sur l’enveloppe. Lis-le bien... et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-4", "key": "d3fab0ead1fc9dcf899f59bddb0c9d4b54373ba1829b6dc49f77fc4027ce1740", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Intense Malevolence Malice. 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": "b5a5333645f965746533792697ee4f426977815db302ffbaa55aacb47615153e", "generation_seconds_batch": 19.97884456103202, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10614672303199768, "target_z": -0.33203123888498176, "ce": 6.078212261199951, "genuineness": 4.68307638168335, "identity": 0.6754622459411621, "hypothesis": "J'ai écrit le numéro sur l'enveloppe, lis le bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d3fab0ead1fc9dcf899f59bddb0c9d4b54373ba1829b6dc49f77fc4027ce1740.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "d404df00775f8fc125638eb57b7988faecb017984de96ed4a79ff6b4d59631f6", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "70adeffc9235fc4857f887687a7be4ca33df5e5bfad9eb4436e9f8a1404fb848", "generation_seconds_batch": 29.214904136955738, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0901013612747192, "target_z": 1.546875001778403, "ce": 6.213237762451172, "genuineness": 4.218649387359619, "identity": 0.47558867931365967, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d404df00775f8fc125638eb57b7988faecb017984de96ed4a79ff6b4d59631f6.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "d430149a91ff9e9ede69657910508eb15e61a4bb63c721f9ca57f1a77e4f545d", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "53b7e90215ae969c16bfbce273f3f30f253e0b06f317065d24c215b9b7a33f5d", "generation_seconds_batch": 26.28892772900872, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0038645267486572266, "target_z": -0.5273437557798095, "ce": 5.9431867599487305, "genuineness": 4.074300289154053, "identity": 0.6741864681243896, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d430149a91ff9e9ede69657910508eb15e61a4bb63c721f9ca57f1a77e4f545d.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "d477e0bbae40d6cd087ddc507d9a88e974d60f64831e7d9d925b2eeb5935e580", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "5fa6f1fb4eab4961f926fdd444aa557325375c09b0e9287164800dd8a4b5f139", "generation_seconds_batch": 26.110362123930827, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8773543834686279, "target_z": 1.1406249484263153, "ce": 6.173524379730225, "genuineness": 4.231201171875, "identity": 0.3839384615421295, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d477e0bbae40d6cd087ddc507d9a88e974d60f64831e7d9d925b2eeb5935e580.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "d5c70abc4181ff06cba16102cf9776b69170d5d8d84f9a299d4a61f1185bcf2a", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "14a6bfbf194146d997cf04514b976fc0945a6d59c49e44c63afc246e982cd70e", "generation_seconds_batch": 27.197832336067222, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01045498251914978, "target_z": -0.5546875115596189, "ce": 6.23706579208374, "genuineness": 4.519899368286133, "identity": 0.5220215320587158, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d5c70abc4181ff06cba16102cf9776b69170d5d8d84f9a299d4a61f1185bcf2a.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "d68ca831031d6558658b973e7f27fb50dd82ac852b486fbea693889a4e813375", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "8df3523f69213261379b4bed2ae80ed2414b616d15126aeabe91a16263d48ccb", "generation_seconds_batch": 27.11878490296658, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0059101879596710205, "target_z": -0.5234374724347548, "ce": 6.23706579208374, "genuineness": 3.860914945602417, "identity": 0.5060828924179077, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d68ca831031d6558658b973e7f27fb50dd82ac852b486fbea693889a4e813375.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "d697771368ac478f34e0912a4cbd7c4457042904d34af1c13490cc4818036a6b", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "26956a645667eda1c2ef119ed788c5e84ed728eafd7fda1714bd614d74116c21", "generation_seconds_batch": 25.06534181407187, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.018184036016464233, "target_z": -0.5, "ce": 6.157639026641846, "genuineness": 4.406930446624756, "identity": 0.7451983094215393, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d697771368ac478f34e0912a4cbd7c4457042904d34af1c13490cc4818036a6b.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": "d6cc12cb081124180e636b34deac7e890b0bdca54c90f25c0a74a4a8d2a65a74", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Escribí el número en el sobre.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b70b1f9fe0347df71d2fd73350b405d88aa50ae148f83358886c909c028a0d76", "generation_seconds_batch": 17.214052723022178, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02022966742515564, "target_z": -0.49609377356383866, "ce": 6.260893821716309, "genuineness": 4.432034492492676, "identity": 0.7106922268867493, "hypothesis": "Escribí el número en el sobre. Léelo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d6cc12cb081124180e636b34deac7e890b0bdca54c90f25c0a74a4a8d2a65a74.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "d6fc42480cd777990d33a5fc679d71c975dacabff9e26c4aaa8edfbd962d2abf", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "d5f4d5200961788679777605c5b08ae7e96bcafda9f44318c9ce3c7ea2df8315", "generation_seconds_batch": 27.197832336067222, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.004318058490753174, "target_z": -0.5429687753422415, "ce": 6.2767791748046875, "genuineness": 4.331617832183838, "identity": 0.612860918045044, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d6fc42480cd777990d33a5fc679d71c975dacabff9e26c4aaa8edfbd962d2abf.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "d70b9bcc2c019447539872ccc544c76abd5ee4211c94d254d1cb972280963443", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "2e41b50affb8a0e2b5826702298bd4f261b83895f94c45b58c765137d844d726", "generation_seconds_batch": 26.500971652916633, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.004318058490753174, "target_z": -0.5429687753422415, "ce": 5.879645347595215, "genuineness": 4.243753433227539, "identity": 0.6088961958885193, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d70b9bcc2c019447539872ccc544c76abd5ee4211c94d254d1cb972280963443.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "d7890f52a7c095e1979980b059bb6b8902852c5e317730298e28084b110b0238", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "a5b49b3b616e22dee2f4d28f395ebd92da7cfe314b376b11bdef9099cf0d092a", "generation_seconds_batch": 25.093500136979856, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03454919159412384, "target_z": -0.4687499893295825, "ce": 5.700935363769531, "genuineness": 4.005263805389404, "identity": 0.34709233045578003, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d7890f52a7c095e1979980b059bb6b8902852c5e317730298e28084b110b0238.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "d8aa7340a56c69e7aceaabf35ac4b990db36686ff8c2d711a3a832f8d2fb3063", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "9efc66483f807d871778c5a1b27e761e67dfa4eb290e55b6967332fb8721236e", "generation_seconds_batch": 19.285099836066365, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00227239727973938, "target_z": -0.5390624919971869, "ce": 6.292664527893066, "genuineness": 4.532451152801514, "identity": 0.5114520192146301, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d8aa7340a56c69e7aceaabf35ac4b990db36686ff8c2d711a3a832f8d2fb3063.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "d8c100f4acf06e2b587428d3557520e0a8d3582e08e2c8155059350951f4d333", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a82ccaef38ceba0f066d2be991b2ce9300a7f6bad39b45ea111f4fabf6133edb", "generation_seconds_batch": 28.571292817941867, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.0329570472240448, "target_z": -0.5976562299929672, "ce": 6.09409761428833, "genuineness": 4.457139015197754, "identity": 0.5309813022613525, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d8c100f4acf06e2b587428d3557520e0a8d3582e08e2c8155059350951f4d333.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": "d8d2f85dbaa89b80b6c3aa4231c4416f03e12e508b4e2286c3fd4eb8dfc59adb", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7a1791e719e90eb27631675fe90ccbdf0d7b1c49d8cb6de0f2133f4ccd982d6f", "generation_seconds_batch": 20.107833947986364, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03148071467876434, "target_z": -0.47460938589271784, "ce": 6.09409761428833, "genuineness": 4.432034492492676, "identity": 0.6486231088638306, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue. Elle était ici, près de la grand bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d8d2f85dbaa89b80b6c3aa4231c4416f03e12e508b4e2286c3fd4eb8dfc59adb.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "d8d85a280a881e3d0b9f58e1b764f8e1d620524d34ca68707515cddd82644639", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "724d9ed5fe69fd795b30ea3aac775ecb9873a8049eaedbd10e933d9e694ebba7", "generation_seconds_batch": 19.52777700603474, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1228315830230713, "target_z": 1.609374852392558, "ce": 6.125868320465088, "genuineness": 3.7291181087493896, "identity": 0.35706058144569397, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d8d85a280a881e3d0b9f58e1b764f8e1d620524d34ca68707515cddd82644639.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "d919f569e5f804b40cf8c38242998c9e99539b2c7d36108cf21ed84d46d813c7", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "4a60f80d46e267d21cae71d13a4d78fb1ed1fad9f1fd72f41deddf3cc4e095b8", "generation_seconds_batch": 23.972444336046465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.016138404607772827, "target_z": -0.5039062264361613, "ce": 6.149696350097656, "genuineness": 4.570107460021973, "identity": 0.4612290561199188, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d919f569e5f804b40cf8c38242998c9e99539b2c7d36108cf21ed84d46d813c7.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": "d96cdedbfd89e6e1d25e43774cdfe99d657dc0d89dc864e560cf47db70e5a4b0", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c1755eaa7b38ab661108513b7299c23837e9c4ea67fd2767ff0eae8a8d6b9107", "generation_seconds_batch": 18.91142566106282, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7791634798049927, "target_z": 0.9531249413127038, "ce": 6.23706579208374, "genuineness": 4.482243061065674, "identity": 0.5654239654541016, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, lies sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d96cdedbfd89e6e1d25e43774cdfe99d657dc0d89dc864e560cf47db70e5a4b0.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "d9ad2c43ec79107b1cf604a2dfdc5a141717ea334360516fdd1a7371873090ee", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "851e469e75102b4de017fda7d463d09ee646b1181dd7aec8266a710fca2287ab", "generation_seconds_batch": 23.933765553985722, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.004318058490753174, "target_z": -0.5429687753422415, "ce": 6.157639026641846, "genuineness": 4.331617832183838, "identity": 0.7258822917938232, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/d9ad2c43ec79107b1cf604a2dfdc5a141717ea334360516fdd1a7371873090ee.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "da13cb5011b78b5a6a506c5026475e6cb9910665a8f23d83684f252cf088979e", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cfc6d1cc0dc4830302339ff17811252563287cbd1c589332f110a202a85ae858", "generation_seconds_batch": 26.08786106808111, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.06364172697067261, "target_z": -0.6562500248976408, "ce": 6.443574905395508, "genuineness": 4.168440818786621, "identity": 0.32942354679107666, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/da13cb5011b78b5a6a506c5026475e6cb9910665a8f23d83684f252cf088979e.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "da237f64c9cf87c9b6aee5aff4da3904b1fd7e7ecb4c73f0c4696db7454b7f2e", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "2142c8e7abff5fc6984a855f9e9effb2f89491f8bc94f333c444bace22ecd23d", "generation_seconds_batch": 24.53881947300397, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9673627614974976, "target_z": 1.3125000497952817, "ce": 6.23706579208374, "genuineness": 4.111956596374512, "identity": 0.5412484407424927, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/da237f64c9cf87c9b6aee5aff4da3904b1fd7e7ecb4c73f0c4696db7454b7f2e.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "da31cc6428510ed65f082103f2f1d543354fab14245fe1d8e4cc38d775787e1e", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "8a3896249fdd98b256129eab5ead0ac15039871238442e2e36155686ac3ab575", "generation_seconds_batch": 30.528664504061453, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.0329570472240448, "target_z": -0.5976562299929672, "ce": 6.459460258483887, "genuineness": 4.381826400756836, "identity": 0.23777492344379425, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/da31cc6428510ed65f082103f2f1d543354fab14245fe1d8e4cc38d775787e1e.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": "dafad78eb03c00fbb5e2bf2a7bd9b8f82eb6060569e6b4d6b4c568f936f0a90b", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4c7c1a9a56f2a14bcae95cd466f89fa71e9bd2781c0c1666a5f7594a261cf496", "generation_seconds_batch": 17.20423267094884, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.026366621255874634, "target_z": -0.48437498043756794, "ce": 6.300607204437256, "genuineness": 4.65797233581543, "identity": 0.786128044128418, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí, junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/dafad78eb03c00fbb5e2bf2a7bd9b8f82eb6060569e6b4d6b4c568f936f0a90b.mp3" }, { "id": "heldout-en-0", "language": "en", "sentences": [ "Your message arrived after I had left.", "I read it twice... before putting the phone away.", "Let me know when we can talk." ], "text": "Your message arrived after I had left. I read it twice... before putting the phone away. Let me know when we can talk.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-0", "key": "db192910e9ae67de251fb2a990940eab881f66f627c78b50b638ce61f5a63679", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Your message arrived after I had left.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Your message arrived after I had left.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Intense Malevolence Malice. 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": "23e3be78adf12a69d9b3804b37c3d44e2654ce51233096786ddc4238a2e23929", "generation_seconds_batch": 20.21946391707752, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0018188953399658203, "target_z": -0.5312499822159709, "ce": 6.467403411865234, "genuineness": 4.118232727050781, "identity": 0.4385647177696228, "hypothesis": "Your message arrived after I had left. I read it twice, before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/db192910e9ae67de251fb2a990940eab881f66f627c78b50b638ce61f5a63679.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "db2b2f0b180157cf64714835cb66e538d78eb995498120c1f66c4bbdc8fb5984", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "35e24c66160a06bbf844b381de041c28ad9a940bcd296c4ac13cc54b7c152821", "generation_seconds_batch": 28.461921441950835, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.018184036016464233, "target_z": -0.5, "ce": 5.796247482299805, "genuineness": 4.319066047668457, "identity": 0.14017021656036377, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/db2b2f0b180157cf64714835cb66e538d78eb995498120c1f66c4bbdc8fb5984.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "dc549d7298185035161adf21f1f430ba94e2ff4d41819a6137d4f77b220eed55", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "b15b241b1ab587d1a73929937d4dbddf065d33942da56db68ac3bcfc8e22fbe2", "generation_seconds_batch": 20.672593241906725, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.026820123195648193, "target_z": -0.5859374937755898, "ce": 6.0146708488464355, "genuineness": 4.017816066741943, "identity": 0.428894579410553, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/dc549d7298185035161adf21f1f430ba94e2ff4d41819a6137d4f77b220eed55.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "dd105944f7ceb3149e962493ce9642a9ea311387f3bbf5795a672c1a523f5a8f", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "f396a8a158512d85c6b748bb2fd7f88e9548c518dec5bb810a565c29ee230372", "generation_seconds_batch": 25.160550824017264, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05091433227062225, "target_z": -0.43750000711361164, "ce": 6.332377910614014, "genuineness": 4.369274139404297, "identity": 0.3875635862350464, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/dd105944f7ceb3149e962493ce9642a9ea311387f3bbf5795a672c1a523f5a8f.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "dd82b8c202de8d0e8006b1973abec0ace0065ae306454400cd6269131b14dd72", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "2c78843211e7d91a2bf4181634c973c04d2f7a3dfd6eb9c7ac08daa1e93f5560", "generation_seconds_batch": 23.136293446994387, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.05341348052024841, "target_z": -0.6367187219901541, "ce": 6.205295085906982, "genuineness": 3.7667744159698486, "identity": 0.2752910852432251, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/dd82b8c202de8d0e8006b1973abec0ace0065ae306454400cd6269131b14dd72.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "ddb1f37d6f8105d49444dd668cc7d5c261965b3d8e7f085bf0c37a76cec03335", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "b52beec86efd912ece65bc29d4c5437d326f3ef1ff1365cf4657e6b26c91ea36", "generation_seconds_batch": 25.15201999794226, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00636368989944458, "target_z": -0.5468750017784029, "ce": 6.157639026641846, "genuineness": 3.9550557136535645, "identity": 0.3283887505531311, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ddb1f37d6f8105d49444dd668cc7d5c261965b3d8e7f085bf0c37a76cec03335.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "ddcd7c44900039ff71f60fd2f522c7ed019321ad009afb4aabfaf9c357d3e0d9", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "7d3952c430741985715cc0b50a8e4fb34e747202b47861de534456cb541c4ecb", "generation_seconds_batch": 27.115666162921116, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04886868596076965, "target_z": -0.4414062620042197, "ce": 6.173524379730225, "genuineness": 4.444586753845215, "identity": 0.48041266202926636, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ddcd7c44900039ff71f60fd2f522c7ed019321ad009afb4aabfaf9c357d3e0d9.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "ddcf9f719da0d1494aefbdda1b3528b6cedab9ab44fe77880b565892a516f161", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "6e1a0c42180dfe8569a8f7e27ad5fb1c70edb99504f12f6461f6bdbf4636a499", "generation_seconds_batch": 18.946993000921793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.010001450777053833, "target_z": -0.5156250195624321, "ce": 6.300607204437256, "genuineness": 4.231201171875, "identity": 0.6716411113739014, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ddcf9f719da0d1494aefbdda1b3528b6cedab9ab44fe77880b565892a516f161.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "ddd8628b978420526d91e79722f4a9f4368670d94f6a0c103117f398ecc29ee8", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d1d8694c0463c90fd18edc57951e7beb7eae52ce9422c52b81a28a8e866b3d16", "generation_seconds_batch": 27.355432596057653, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00227239727973938, "target_z": -0.5390624919971869, "ce": 6.117925643920898, "genuineness": 4.400654315948486, "identity": 0.5762283802032471, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ddd8628b978420526d91e79722f4a9f4368670d94f6a0c103117f398ecc29ee8.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "de06e00cb288886ab4db34d4621b1a63fd8b604aec878387e8dd211397a099ec", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "45368b84a459927604c36f2f85651225b2dc3f59782d96973ce26ce74632a4c6", "generation_seconds_batch": 26.69307792407926, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.007955819368362427, "target_z": -0.5195312459985935, "ce": 6.1894097328186035, "genuineness": 4.406930446624756, "identity": 0.6470088958740234, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/de06e00cb288886ab4db34d4621b1a63fd8b604aec878387e8dd211397a099ec.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "de1247028a250fea611f134bb3ba64f6d5903da3148a918d6482d663af022e46", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "f86b9c1bda6ba144a65c2673a1e6ad60afb7d2f0a93baf3edad9162ddfa04718", "generation_seconds_batch": 20.977512757992372, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.023298144340515137, "target_z": -0.49023437700070327, "ce": 5.77639102935791, "genuineness": 3.810706615447998, "identity": 0.3936226963996887, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/de1247028a250fea611f134bb3ba64f6d5903da3148a918d6482d663af022e46.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "de46fc429566e9ae46e10400013eda959f741c26764b483227957499990f20df", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e482658c25b6ba663dc7f8c27d24bd010f22b64b40915ac14337e905f34ba3e6", "generation_seconds_batch": 27.862253987928852, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01454627513885498, "target_z": -0.562500021340835, "ce": 6.348263263702393, "genuineness": 4.168440818786621, "identity": 0.399832159280777, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/de46fc429566e9ae46e10400013eda959f741c26764b483227957499990f20df.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "de94db47297ee1073413f81eb3ce5a908537c61b87c9559ed56b086bf90ecbba", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "c4762eec4af4752b0ce95d226287c18d35ecc3a6d26c6552175bcf01ed4b8bf1", "generation_seconds_batch": 27.93442015699111, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.227351114153862, "target_z": -0.10058592827453487, "ce": 6.173524379730225, "genuineness": 3.9550557136535645, "identity": 0.1906222403049469, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/de94db47297ee1073413f81eb3ce5a908537c61b87c9559ed56b086bf90ecbba.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "dee89da52d1244794d478b1537cf1152dd9b35f9ccc8307e9c3c27945af139eb", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "34acb1c427171febab2190e084e659e9b3aef9be120398955356fbc7a8496400", "generation_seconds_batch": 23.911939401063137, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01045498251914978, "target_z": -0.5546875115596189, "ce": 6.141753673553467, "genuineness": 4.507347106933594, "identity": 0.6870912313461304, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/dee89da52d1244794d478b1537cf1152dd9b35f9ccc8307e9c3c27945af139eb.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "df062d3a3151c8c44e7095444777f52853e2038f5ddc7f7f41c4719b3f9fb1d1", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "bc3e69e6fe302bc747cf19e4d380642c3c9c03c409c4da8f5142bd966fccdfba", "generation_seconds_batch": 26.740647199912928, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.041708946228027344, "target_z": -0.45507811143967775, "ce": 6.133810997009277, "genuineness": 4.570107460021973, "identity": 0.37880536913871765, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/df062d3a3151c8c44e7095444777f52853e2038f5ddc7f7f41c4719b3f9fb1d1.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "df1b5b3b36d2f5d860fc0735510ec287046e3753aee2701959b00ac5abb581dc", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "92fb1e82948804111de4178d6821adc54294cdd5711e1fe7cce91f141841c6f8", "generation_seconds_batch": 21.528846655972302, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04068611562252045, "target_z": -0.4570312531122051, "ce": 6.038498878479004, "genuineness": 4.287685871124268, "identity": 0.7331370115280151, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/df1b5b3b36d2f5d860fc0735510ec287046e3753aee2701959b00ac5abb581dc.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": "df488d474533f440268ee397460f1af6c6600d6d0bb6ebb289c6cd8c9e343142", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The keys are on the kitchen table.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3f5e6ceb13732f53c6ac1a4765f9806bca19bc9209b8214642ceb1660bee0464", "generation_seconds_batch": 26.662643237039447, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.05750477313995361, "target_z": -0.6445312317713701, "ce": 6.38003396987915, "genuineness": 4.319066047668457, "identity": 0.3783237934112549, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you leave.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/df488d474533f440268ee397460f1af6c6600d6d0bb6ebb289c6cd8c9e343142.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "df6da2c403fa0059af5fdb246e812f51e22fa29d493db1deab0e6592f3aa8e1a", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "43c4be559e732ad189fa6c20f2e7602f150fcc9478eb89ce319f918b97929c74", "generation_seconds_batch": 27.98958996508736, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09898696839809418, "target_z": -0.3457031167748865, "ce": 6.1894097328186035, "genuineness": 3.8922953605651855, "identity": 0.488228976726532, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/df6da2c403fa0059af5fdb246e812f51e22fa29d493db1deab0e6592f3aa8e1a.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "dffdb64ac5ffad61af994b78551e69301fb8338717f1e57ff6d1e854fab9d88d", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "651f7a4b9a16ab98ba6b3abb5dbb99c3add3dd1adf5450da4aeb5a9b1bff4b76", "generation_seconds_batch": 20.780900493031368, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0059101879596710205, "target_z": -0.5234374724347548, "ce": 6.125868320465088, "genuineness": 3.9425034523010254, "identity": 0.48407888412475586, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/dffdb64ac5ffad61af994b78551e69301fb8338717f1e57ff6d1e854fab9d88d.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "e0029f39d2fe0472b590149159c5dfc6ec0dd9843e19b73dae1713c3cf87b78f", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "e506d226a8bd1f52c0b3127f431c5aa462b8433d3b7ad8892adfb60cce252c30", "generation_seconds_batch": 29.214904136955738, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.225113868713379, "target_z": 1.8046875400140656, "ce": 6.213237762451172, "genuineness": 4.243753433227539, "identity": 0.3600575923919678, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e0029f39d2fe0472b590149159c5dfc6ec0dd9843e19b73dae1713c3cf87b78f.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "e117261e88fc19ea9ce594bc8eb78e4b3c72ca13387957a671d93ad5fe030f13", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "3b79d132a24b7f91eb1e9f082950526a2837153f66fc811d451a302fe5f9b544", "generation_seconds_batch": 26.237071320065297, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8937195539474487, "target_z": 1.1718749875511796, "ce": 6.324435234069824, "genuineness": 3.9927117824554443, "identity": 0.290377676486969, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e117261e88fc19ea9ce594bc8eb78e4b3c72ca13387957a671d93ad5fe030f13.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": "e147bcfe641f0cb00ee0ced0ba9659363a3fcf00fd3cf589aa4c423b4d841bcb", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8fe516c0ae0499ef2071deea64eefe5fec5d361e3438aa7c2afd8de0534ad091", "generation_seconds_batch": 17.20423267094884, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03659482300281525, "target_z": -0.46484376289342116, "ce": 6.078212261199951, "genuineness": 4.457139015197754, "identity": 0.6729182004928589, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí, junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e147bcfe641f0cb00ee0ced0ba9659363a3fcf00fd3cf589aa4c423b4d841bcb.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "e1715e8a4d7951e5c0c50313537d23edd4ee244513ff4abde4b11e73f76c7cf0", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "4a5cf96d350ff0401b38ef2ad6cc204b405857afa01334055f44b1e1f1333ffe", "generation_seconds_batch": 27.757757390034385, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03454919159412384, "target_z": -0.4687499893295825, "ce": 6.062326908111572, "genuineness": 4.419482707977295, "identity": 0.6280707120895386, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e1715e8a4d7951e5c0c50313537d23edd4ee244513ff4abde4b11e73f76c7cf0.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "e188ece9a65581f5078ddfd0152fe91916c2b4b77bd36ca1303039d8cf36d0d5", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "0ef8a6ba09131669e86e7699644ef5b41a5574ade03af6ea1c9bd8c3100aba1b", "generation_seconds_batch": 20.22929605608806, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0038645267486572266, "target_z": -0.5273437557798095, "ce": 6.117925643920898, "genuineness": 4.068024158477783, "identity": 0.4388172924518585, "hypothesis": "I found a little box. It was beside the old window. We can look inside together.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e188ece9a65581f5078ddfd0152fe91916c2b4b77bd36ca1303039d8cf36d0d5.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "e230aac55ce95f40940d2268ebf33e8cb60fad387a13e1fd00d392867e545a70", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "23d146e3cd16204c986e07b0732e46937539f264070e771ca11ae3b2de9c6dcd", "generation_seconds_batch": 24.778106614015996, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.014092743396759033, "target_z": -0.5078125097812161, "ce": 6.213237762451172, "genuineness": 4.319066047668457, "identity": 0.3177282214164734, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e230aac55ce95f40940d2268ebf33e8cb60fad387a13e1fd00d392867e545a70.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": "e2958c7dc47274b69c62274a3b4617008fa016349d55db434c93e29132ad5bd7", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "4e7ef82a32662b04b1ab2d9a9a49f1d6b8fd54bda27128ce3b178ec021154f2c", "generation_seconds_batch": 17.096285445033573, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04273176193237305, "target_z": -0.4531249982215971, "ce": 5.95112943649292, "genuineness": 4.21237325668335, "identity": 0.6195187568664551, "hypothesis": "¿La carretera gira a la izquierda tras el puente? Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e2958c7dc47274b69c62274a3b4617008fa016349d55db434c93e29132ad5bd7.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "e3a000df0dc0a1c8107c6703888fe344a6d8af96d5c1793ac4ee613eedb5c48f", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "73f3b60b7e25a51748522e17278820a177f29fa67ad5e8d304916b99ac261c36", "generation_seconds_batch": 30.710311016999185, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.059096917510032654, "target_z": -0.4218749875511796, "ce": 6.427689552307129, "genuineness": 4.149612903594971, "identity": 0.5565155148506165, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e3a000df0dc0a1c8107c6703888fe344a6d8af96d5c1793ac4ee613eedb5c48f.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "e3df1e747c6e049520dc196fd2ea90237a1a8f7159d56d14e65909f806dcd66d", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ab09c670284d4d8a4f10cdc82dea89046a8adec6937a25707aaad5a243c0bb71", "generation_seconds_batch": 28.599422290106304, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1531965285539627, "target_z": -0.24218749021878397, "ce": 6.046441555023193, "genuineness": 3.7542223930358887, "identity": 0.23169711232185364, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e3df1e747c6e049520dc196fd2ea90237a1a8f7159d56d14e65909f806dcd66d.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "e3e83d2d5d3debde365951723b545231448091a82c25850cd2d0e4ce5a838d8f", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "bfbd775b6474fb05e0a2bb9b9226cfb4218c419bb7914740b68330216d812497", "generation_seconds_batch": 24.73673979495652, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0018188953399658203, "target_z": -0.5312499822159709, "ce": 6.022613525390625, "genuineness": 4.306513786315918, "identity": 0.7344553470611572, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e3e83d2d5d3debde365951723b545231448091a82c25850cd2d0e4ce5a838d8f.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": "e445a999bec9e5ac2a06e2ccadd0a61bc778f20fce87283b9b311e9f8e1c1eba", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6dbedc54686d018c9a939d343d1b2e63311fde7bbe02452beffecd7fce63f7a3", "generation_seconds_batch": 18.71437037095893, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02841225266456604, "target_z": -0.48046875400140654, "ce": 6.022613525390625, "genuineness": 4.218649387359619, "identity": 0.6888972520828247, "hypothesis": "Las llaves están en la mesa de la cocina. Miré en el cajón y luego, junto a la ventana, llévatelas cuando te vayas.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e445a999bec9e5ac2a06e2ccadd0a61bc778f20fce87283b9b311e9f8e1c1eba.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "e455aa330e9e4101b597811a0e335b6f081a2e4f4625196a8b56f037117d9900", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "f0634cc791334a4c21c2263d44c255ae4fa502d82d0e501ddb8c8a488cd800f9", "generation_seconds_batch": 19.40704492200166, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03250354528427124, "target_z": -0.47265624422019054, "ce": 6.221180438995361, "genuineness": 4.532451152801514, "identity": 0.1790710687637329, "hypothesis": "La porte est encore ouverte. J'ai posé ton anteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e455aa330e9e4101b597811a0e335b6f081a2e4f4625196a8b56f037117d9900.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "e46ae609e90a59389bfbe0ade53193d894700b2d72d6e102b49e8012a6f7abbb", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "16c743fea128cc58d608425ad1c71596dea009f59e517bb0200c7f9407333bb3", "generation_seconds_batch": 26.384779763990082, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08159899711608887, "target_z": -0.3789062406633847, "ce": 6.284721851348877, "genuineness": 4.111956596374512, "identity": 0.428225040435791, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e46ae609e90a59389bfbe0ade53193d894700b2d72d6e102b49e8012a6f7abbb.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": "e50c7504ade66c0dd1d3cfd6819df18e4714c7ecd3ccfbeba69699109a799954", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Escribí el número en el sobre.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2ac2340bb6cfc8b2fc91b86831071d5e191530ba4e4faab0b6b73e899f4b72c6", "generation_seconds_batch": 23.035625742049888, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04477740824222565, "target_z": -0.44921874333098905, "ce": 6.197352409362793, "genuineness": 4.118232727050781, "identity": 0.5581135749816895, "hypothesis": "Escribí el número en el sobre, léelo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e50c7504ade66c0dd1d3cfd6819df18e4714c7ecd3ccfbeba69699109a799954.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": "e50d11b1a6264f40dd63ca0540aacb43039fcc998af042b78b02cdb1bef813de", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4b5c6f8fc3548891712eb6096cb5479041dbae200152c1eb31c50c1015f8cdfb", "generation_seconds_batch": 22.437216124031693, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10307824611663818, "target_z": -0.33789063544811715, "ce": 6.149696350097656, "genuineness": 4.469690799713135, "identity": 0.606175422668457, "hypothesis": "J'ai écrit le numéro sur l'enveloppe. Lis-le bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e50d11b1a6264f40dd63ca0540aacb43039fcc998af042b78b02cdb1bef813de.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": "e50dad2dfdf752981019882d00199805c2db3d1c5679ca51d7bc5d7f3bfdbb2a", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "517eb8458ab1680835d945355d3d74544fe59adddf9f091ba1a0b982237e4a71", "generation_seconds_batch": 17.1304574070964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.0411396324634552, "target_z": -0.6132812495553993, "ce": 6.24500846862793, "genuineness": 3.9927117824554443, "identity": 0.6648598909378052, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e50dad2dfdf752981019882d00199805c2db3d1c5679ca51d7bc5d7f3bfdbb2a.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "e53104f756595ac13246ae5665e8a6f69ae9095e103d571f7033aad0fc376386", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "e7bfa251c3f9ae4b1dc14d82a26bf8edd2cf6b8d5bdc171c4dea60548983c61c", "generation_seconds_batch": 20.977512757992372, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02022966742515564, "target_z": -0.49609377356383866, "ce": 5.8875885009765625, "genuineness": 3.948779582977295, "identity": 0.6738898754119873, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e53104f756595ac13246ae5665e8a6f69ae9095e103d571f7033aad0fc376386.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "e541587fe437b16fea154bf8864dc5d4b0a96924b9932952dce808088a72cd06", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "c5b414589034840c70fdf881c60cbc700c6818f24a77563d5951ec72c3df6ff6", "generation_seconds_batch": 26.650074403965846, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.16802743077278137, "target_z": -0.2138672062843808, "ce": 6.086154937744141, "genuineness": 4.306513786315918, "identity": 0.679164469242096, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e541587fe437b16fea154bf8864dc5d4b0a96924b9932952dce808088a72cd06.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "e58187e83d1053ba69b7d688033d246baba93338da7d2b64839765c800c63faa", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "e793237dac4872f7e81be6a4a9ae793b9c25449bad98dfde76c8193d230a13a7", "generation_seconds_batch": 29.03429691796191, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2765746116638184, "target_z": 3.812499765250815, "ce": 6.1894097328186035, "genuineness": 3.9801597595214844, "identity": 0.24579980969429016, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e58187e83d1053ba69b7d688033d246baba93338da7d2b64839765c800c63faa.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": "e5d452c4d5b85f1c84db561fe97590137f066cde4b7e8407084bc3a67ac74bc9", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b7697aed48af0b2f311e5b97e2622e2f9bac32552fb6ca947c12371d1ab47159", "generation_seconds_batch": 19.090801777900197, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05193716287612915, "target_z": -0.43554686544108434, "ce": 6.062326908111572, "genuineness": 4.068024158477783, "identity": 0.5683017373085022, "hypothesis": "J'ai écrit le numéro sur l'enveloppe. Lis-le bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e5d452c4d5b85f1c84db561fe97590137f066cde4b7e8407084bc3a67ac74bc9.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "e5e3f3a433ff811dffc7bfe16e5d8edfd2c8e31298d8a90b1410fee542ac78ad", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "31686dffc0ed47422cb8eb3ffedcef1f917f7f32bb5733173723c4694d2cb548", "generation_seconds_batch": 21.796908541000448, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.016591906547546387, "target_z": -0.5664062477769963, "ce": 6.260893821716309, "genuineness": 4.356721878051758, "identity": 0.6690660119056702, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e5e3f3a433ff811dffc7bfe16e5d8edfd2c8e31298d8a90b1410fee542ac78ad.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": "e5e99c4d885d3d82a741ba9b6ccf08e3239f4895c8ad0ee3c0e76ab9ed7f9e23", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b0b57eb5c291b486b6bd30a7a923d8ddab1156c7872a88538a143bd5bbc978a4", "generation_seconds_batch": 23.243263807031326, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.012500613927841187, "target_z": -0.5585937379957803, "ce": 6.2767791748046875, "genuineness": 4.369274139404297, "identity": 0.5856825709342957, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis moi, quand nous pourrons parler?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e5e99c4d885d3d82a741ba9b6ccf08e3239f4895c8ad0ee3c0e76ab9ed7f9e23.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": "e6676aa11373e11e67e024d8131c9ec5a25694b17e73db23c917457847b9732f", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "85a9ce860f0c07d1c53345eb97331a0b19219514a87a5d892c63e51e9cb62148", "generation_seconds_batch": 23.347755653085187, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.024774491786956787, "target_z": -0.5820312673394284, "ce": 6.205295085906982, "genuineness": 4.15588903427124, "identity": 0.6955072283744812, "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/emo_Malevolence_Malice/e6676aa11373e11e67e024d8131c9ec5a25694b17e73db23c917457847b9732f.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": "e675c3494183acb08467f7c582b8490f1aac40d5a5072f95722a3f6e93ed64df", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c703b103216153866eab9aa8db2b53916db1652d593ad86c97d78f391a6c4a84", "generation_seconds_batch": 17.19549462106079, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5684621930122375, "target_z": 0.5507812282145643, "ce": 6.24500846862793, "genuineness": 4.21237325668335, "identity": 0.4248064160346985, "hypothesis": "Ich schrieb die Nummer auf den Umschlag. Liese genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.09090909090909091, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e675c3494183acb08467f7c582b8490f1aac40d5a5072f95722a3f6e93ed64df.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "e68f723b0a5022d1eb32dfd0beebe487155f3e931437d3299914e30c122a829c", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "2a56b10e72c04898ffe8e310ab174f9c42ee5e58413751b2f0cf3f0a028fe6fe", "generation_seconds_batch": 25.179453340009786, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0038645267486572266, "target_z": -0.5273437557798095, "ce": 6.117925643920898, "genuineness": 4.425758361816406, "identity": 0.7461338043212891, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e68f723b0a5022d1eb32dfd0beebe487155f3e931437d3299914e30c122a829c.mp3" }, { "id": "heldout-en-3", "language": "en", "sentences": [ "Someone moved the little blue chair.", "It used to stand here... beside the tall bookshelf.", "We can put it back after lunch." ], "text": "Someone moved the little blue chair. It used to stand here... beside the tall bookshelf. We can put it back after lunch.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-3", "key": "e6f1e80cc6606ef00fdc322d5829b94fd65ad80a74ab015f88321e5dc194804c", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Someone moved the little blue chair.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Someone moved the little blue chair.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Intense Malevolence Malice. 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": "f42784e1cf124141ab9497ada73ba3180d9ce24d3d347e3263dc975519b5bcef", "generation_seconds_batch": 23.75255783495959, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00227239727973938, "target_z": -0.5390624919971869, "ce": 6.340320587158203, "genuineness": 4.325342178344727, "identity": 0.4062601327896118, "hypothesis": "Someone moved the little blue chair. It used to stand here beside the tall bookshelf. We can put it back after lunch.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e6f1e80cc6606ef00fdc322d5829b94fd65ad80a74ab015f88321e5dc194804c.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": "e7783f7283dc01c671d3f5318a9e72aeed41636a16e1f62c567e6086e4d6ada7", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e866c1ddb9d484488c005647abb29ea1210203fe70e0f7fb92ed0ccda3dca134", "generation_seconds_batch": 27.54404737695586, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00227239727973938, "target_z": -0.5390624919971869, "ce": 6.41180419921875, "genuineness": 4.344170093536377, "identity": 0.6695765852928162, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter, bis zum kleinen Holztor. Ich warte auf der anderen Seite.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e7783f7283dc01c671d3f5318a9e72aeed41636a16e1f62c567e6086e4d6ada7.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "e7b61d1678a9c2459f30898b8c09644079acba13f41cd1cbab5c1f7e41169f61", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "c27ee0dbda6f082122fda97c8732f91cf8d116a8baf690c5c861ad0d3f23d311", "generation_seconds_batch": 19.91579747700598, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5101613402366638, "target_z": 0.4394530918772457, "ce": 6.308549880981445, "genuineness": 4.2374773025512695, "identity": 0.06401152163743973, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e7b61d1678a9c2459f30898b8c09644079acba13f41cd1cbab5c1f7e41169f61.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "e891d72c2d445da45a63b925782103279814a7a1fdd956e501d6ef47b2d46cc6", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "95cd3631a131b14f3e66cd34de1ab6e009d8174ebcbdb9256ff13724cba0e9fb", "generation_seconds_batch": 29.262115109944716, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03148071467876434, "target_z": -0.47460938589271784, "ce": 6.141753673553467, "genuineness": 4.168440818786621, "identity": 0.762651801109314, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise, dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e891d72c2d445da45a63b925782103279814a7a1fdd956e501d6ef47b2d46cc6.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "e93895c01051d5ea22db1d5352dc6515b5b3c858bfbb30566e42ef913fcac5b1", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "bfd22adcb52d41385f28ed9969dfa607b79fc3d42478b0a4231483eac8221dee", "generation_seconds_batch": 20.72498626401648, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.23911356925964355, "target_z": -0.07812498399437376, "ce": 6.006728172302246, "genuineness": 3.7793264389038086, "identity": 0.23927150666713715, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e93895c01051d5ea22db1d5352dc6515b5b3c858bfbb30566e42ef913fcac5b1.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": "e93f4b5ff9a33fac1748985f9de1b59fb554f3e7dd99701374116b26032179e5", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b1fe858b18d5bd5e446c2f8c7d734818cc6603d72b9d702fe0d385ae59213c10", "generation_seconds_batch": 22.320822505047545, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.024774491786956787, "target_z": -0.5820312673394284, "ce": 5.808161735534668, "genuineness": 4.394378662109375, "identity": 0.29042112827301025, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue, elle était ici près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e93f4b5ff9a33fac1748985f9de1b59fb554f3e7dd99701374116b26032179e5.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "e95eb5e40a0ddfaa9710a58bb33d6e3ed25b0e0e7d645d349de442fd68413b30", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "418077883459bf006ae8f16203afc5db502defa0351303564a250f00cd2cda02", "generation_seconds_batch": 29.813897681073286, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0059101879596710205, "target_z": -0.5234374724347548, "ce": 6.181467056274414, "genuineness": 4.632867813110352, "identity": 0.6930654048919678, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e95eb5e40a0ddfaa9710a58bb33d6e3ed25b0e0e7d645d349de442fd68413b30.mp3" }, { "id": "heldout-de-3", "language": "de", "sentences": [ "Jemand hat den kleinen blauen Stuhl umgestellt.", "Er stand hier... neben dem hohen Bücherregal.", "Wir stellen ihn nach dem Essen zurück." ], "text": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier... neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-3", "key": "e9d266c3907611a53d0c5f7688a7e231bba29ead6f87b8ae8d29cb37ab2fae41", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Intense Malevolence Malice. 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": "809b2f707c11014ed79d0818ea42521408772af7816580a7c7dbfb3b85cd29a0", "generation_seconds_batch": 18.544277752982453, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.010001450777053833, "target_z": -0.5156250195624321, "ce": 6.292664527893066, "genuineness": 4.243753433227539, "identity": 0.6143414378166199, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/e9d266c3907611a53d0c5f7688a7e231bba29ead6f87b8ae8d29cb37ab2fae41.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "ea6b8722609c241dd7d7414ca2d1ff0383008d06a0d1a583f5a41273a22e2148", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "984345e6ae47b75bf6992fbda7e584f30b8308cd65847a080f6209914394494e", "generation_seconds_batch": 26.153902292018756, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.04727655649185181, "target_z": -0.6249999857727767, "ce": 5.967014789581299, "genuineness": 4.093128204345703, "identity": 0.5582141280174255, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ea6b8722609c241dd7d7414ca2d1ff0383008d06a0d1a583f5a41273a22e2148.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "eaab30d6635ae93682dd4752a669af8fd947e005e23c7353a14bc8c608d25e42", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "6d7f9e2b0515dd6d41fb8d8effb259881ee82b8c8b7332000a4913956db68781", "generation_seconds_batch": 24.20346673997119, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.010001450777053833, "target_z": -0.5156250195624321, "ce": 6.197352409362793, "genuineness": 3.8922953605651855, "identity": 0.24496352672576904, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/eaab30d6635ae93682dd4752a669af8fd947e005e23c7353a14bc8c608d25e42.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "eaee6ccc1571f6b562f9e587fa5ca4e8185a5b8247d2e0289362aac482578ac4", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "1f139c82bcf3409248e179fcef2cd3fba5c5571c81d2ad5c33e0865a2ed48a13", "generation_seconds_batch": 25.56417757703457, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04068611562252045, "target_z": -0.4570312531122051, "ce": 6.09409761428833, "genuineness": 3.9801597595214844, "identity": 0.6378097534179688, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/eaee6ccc1571f6b562f9e587fa5ca4e8185a5b8247d2e0289362aac482578ac4.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": "eaf6654b714677a7639d60f03c6ff462f996b11010cca94879ff38fde8e478e1", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The road turns left behind the bridge.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b6ef13f888c8b0c62dc8327b10762beecde528bed98e7b4fbda111aec6da92fa", "generation_seconds_batch": 17.17746460007038, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07137078046798706, "target_z": -0.39843748666197815, "ce": 6.372091293334961, "genuineness": 3.8169827461242676, "identity": 0.3063191771507263, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/eaf6654b714677a7639d60f03c6ff462f996b11010cca94879ff38fde8e478e1.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "eba11f4a4a94fbf65c739afbdf05923a4c4d67eb2d2a5149b5a68181c4c5c837", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "126a0f46433171c35b8fe414053d84dbfb5a362f94ab54c129fc0743b6d4d427", "generation_seconds_batch": 25.36149588006083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8200764060020447, "target_z": 1.0312500391248642, "ce": 6.157639026641846, "genuineness": 4.2249250411987305, "identity": 0.3171346187591553, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/eba11f4a4a94fbf65c739afbdf05923a4c4d67eb2d2a5149b5a68181c4c5c837.mp3" }, { "id": "heldout-fr-2", "language": "fr", "sentences": [ "La route tourne à gauche après le pont.", "Continue... jusqu’à la petite barrière en bois.", "Je t’attendrai de l’autre côté." ], "text": "La route tourne à gauche après le pont. Continue... jusqu’à la petite barrière en bois. Je t’attendrai de l’autre côté.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-2", "key": "ebe89981a89e9506d02bacdfa583f9bfe1f04faa699193749ab6419de85c588f", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Intense Malevolence Malice. 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": "7a87d49250c079495fcb6797613dd46a70164b113d47e6b597a9bc709d1c8d20", "generation_seconds_batch": 18.662544685998, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04989151656627655, "target_z": -0.43945312033169237, "ce": 6.24500846862793, "genuineness": 4.507347106933594, "identity": 0.77911376953125, "hypothesis": "La route tourne à gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ebe89981a89e9506d02bacdfa583f9bfe1f04faa699193749ab6419de85c588f.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "ec047febe9f1bb0e03f7e48c1885bcca89bbb64344b53d5cd6cc53d123244529", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "e64961749bc010eee4d3acac64799085b7c3eea8ab5daf5336ad993b06485d0a", "generation_seconds_batch": 18.864009695942514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0901013612747192, "target_z": 1.546875001778403, "ce": 6.070269584655762, "genuineness": 4.281409740447998, "identity": 0.4310005009174347, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ec047febe9f1bb0e03f7e48c1885bcca89bbb64344b53d5cd6cc53d123244529.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "ec0f901126d1c43f82f95bc569c3035ed3047ee6084a8c9624150ecd9ff4e849", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "deb72a33b2ca41b09dd88e1ba53fd6d821454c182d7e31b1484739b7d9f0139d", "generation_seconds_batch": 26.63033886998892, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.03704833984375, "target_z": -0.6054687397741833, "ce": 6.372091293334961, "genuineness": 4.406930446624756, "identity": 0.2690351605415344, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ec0f901126d1c43f82f95bc569c3035ed3047ee6084a8c9624150ecd9ff4e849.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "ec5e34fa4e2e287263708f89d51854a636aa45c785cc3d2f5c12278655f17eee", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "a5a665f1784970730b94222f100b2b5f86659e7794ca2da1664aec8b481aefb1", "generation_seconds_batch": 30.528664504061453, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.020683199167251587, "target_z": -0.5742187575582124, "ce": 6.403861999511719, "genuineness": 4.055472373962402, "identity": 0.2557655870914459, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ec5e34fa4e2e287263708f89d51854a636aa45c785cc3d2f5c12278655f17eee.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "eca7a56ed429e1ff31ebabe25374d4d30321d4d0fe6fd58326388e12f6912685", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "20a4dce3871b191b303b7aa21f2fbb902a61ef3a23dd709ba6db8cf1dbda005b", "generation_seconds_batch": 25.97657679603435, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.21047455072402954, "target_z": -0.13281249555399272, "ce": 6.316492557525635, "genuineness": 4.394378662109375, "identity": 0.5651571750640869, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/eca7a56ed429e1ff31ebabe25374d4d30321d4d0fe6fd58326388e12f6912685.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "ecfe9a42283040ccf5bb0e3265ed5c90038be563e13a4aad0a2124c5bc6df981", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "9643224e2186943dda9e7ec4c99b5f7a6de10f0d5e66fc889cb0f254056e7cba", "generation_seconds_batch": 25.79413764702622, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.026366621255874634, "target_z": -0.48437498043756794, "ce": 5.879645347595215, "genuineness": 3.936227321624756, "identity": 0.4724707007408142, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ecfe9a42283040ccf5bb0e3265ed5c90038be563e13a4aad0a2124c5bc6df981.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": "ed12f372fce7379f5bac62f84b6c83643b5f061b5541d73fd6f448b16970e76b", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dcc25a1715361241e20cb53650963a484ad9ce52a2fd07e00da0b7577af5c1fd", "generation_seconds_batch": 27.54404737695586, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.012500613927841187, "target_z": -0.5585937379957803, "ce": 6.395919322967529, "genuineness": 4.20609712600708, "identity": 0.5791935324668884, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter bis zum kleinen Holztor. Ich warte auf der anderen Seite.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ed12f372fce7379f5bac62f84b6c83643b5f061b5541d73fd6f448b16970e76b.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "ed37bd12839f98dad9a016b3bbf594a4d950b678fa44af940c7e26258e06a638", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "99f4718890adb6c9f796e7835c8f276f73c55daab5257fd158f049160d6ada78", "generation_seconds_batch": 22.552532361936755, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03557200729846954, "target_z": -0.46679687611150184, "ce": 6.078212261199951, "genuineness": 4.507347106933594, "identity": 0.6772094964981079, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ed37bd12839f98dad9a016b3bbf594a4d950b678fa44af940c7e26258e06a638.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "edfd4ddd87fc404ed3ca7ea74f37943c08630b10bf7d46f661ce6862f924b468", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "c96a5b710dc6a57a73503011c9ee599da621a221347634cc1d2136520dd51c1c", "generation_seconds_batch": 26.788008840056136, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09489567577838898, "target_z": -0.3535156265561025, "ce": 6.316492557525635, "genuineness": 4.055472373962402, "identity": 0.4870648980140686, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/edfd4ddd87fc404ed3ca7ea74f37943c08630b10bf7d46f661ce6862f924b468.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "ee880d80542927da18088d82cc203eca7d76624d493ffba6fffb8433daf75eda", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "422ebc2961056b065122aee8fcd33339067f8c9fb9243d5e3db898c3fb32b13f", "generation_seconds_batch": 27.02010803401936, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.016138404607772827, "target_z": -0.5039062264361613, "ce": 6.117925643920898, "genuineness": 4.381826400756836, "identity": 0.5558813214302063, "hypothesis": "De deuren staan nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ee880d80542927da18088d82cc203eca7d76624d493ffba6fffb8433daf75eda.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "eeabfd67bd8784e64998d8c253b0e77c352c9dcfbf48cf2268e426b6210b4068", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "dd2f65238e742b22885de99c5c12eaf6803d1884926552c820c0028b738c5043", "generation_seconds_batch": 24.53881947300397, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.472956418991089, "target_z": 4.187499779478038, "ce": 5.236289024353027, "genuineness": 3.6851859092712402, "identity": 0.3672668933868408, "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/emo_Malevolence_Malice/eeabfd67bd8784e64998d8c253b0e77c352c9dcfbf48cf2268e426b6210b4068.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "ef3258dbc73d07d29bd2454bda201bc65d11bb01ef198dd67dbe9632f66e13ef", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "0980d9421aae05db506b4b5f9d9f3bb6f1f579eb5cf9ad4fde136edae01c8f37", "generation_seconds_batch": 19.966771184001118, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.38946837186813354, "target_z": 0.20898436633028578, "ce": 6.173524379730225, "genuineness": 4.13078498840332, "identity": 0.28885138034820557, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ef3258dbc73d07d29bd2454bda201bc65d11bb01ef198dd67dbe9632f66e13ef.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "ef365b73e1ea6b97c8310f85062bcab4be5f73e7118406d435eea804018b529d", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "23ea809db0462ef4ac6b4c01d4bb4ce337b8277a42fa9c22d3afad9d693b2f21", "generation_seconds_batch": 28.870381204993464, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.026366621255874634, "target_z": -0.48437498043756794, "ce": 6.0146708488464355, "genuineness": 4.21237325668335, "identity": 0.2042786180973053, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ef365b73e1ea6b97c8310f85062bcab4be5f73e7118406d435eea804018b529d.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "ef51f1b22c1bee6347d3443db94ee96407957eefd4e3f3f8acf0382fb31186ce", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "d59b42850796404271463382d5fece2e0e837acb844644219b4804c2f16f3b7c", "generation_seconds_batch": 18.946993000921793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02125251293182373, "target_z": -0.49414060343686467, "ce": 6.268836498260498, "genuineness": 4.607763767242432, "identity": 0.5643211603164673, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ef51f1b22c1bee6347d3443db94ee96407957eefd4e3f3f8acf0382fb31186ce.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": "ef564924646b805d77f4ab265b1ae15e2018a92d85f74f705debd2a75f281fc9", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Someone moved the little blue chair.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5608f7ed1f543f516d77ded3d46e8a49728470fe31064867ee4791ae3bab090f", "generation_seconds_batch": 21.012329844990745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00227239727973938, "target_z": -0.5390624919971869, "ce": 6.38003396987915, "genuineness": 4.293961524963379, "identity": 0.32406920194625854, "hypothesis": "Someone moved the little blue chair. It used to stand here, beside the tall bookshelf. We can put it back after lunch.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ef564924646b805d77f4ab265b1ae15e2018a92d85f74f705debd2a75f281fc9.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "ef7b250bde865a74bd192e99d42ca915ee00c43ea488b4effcf91f9c7c7d2ffa", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fe351cf58d9b74b37ad04dfe4892f034fe19c9d57dbd99af1770ca2a9c9c66b2", "generation_seconds_batch": 27.479947610991076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.42424431443214417, "target_z": 0.27539061410728216, "ce": 6.070269584655762, "genuineness": 4.118232727050781, "identity": 0.48637643456459045, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ef7b250bde865a74bd192e99d42ca915ee00c43ea488b4effcf91f9c7c7d2ffa.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "f0901d327c16fea849ba5fac948e1ee41ff567a650dc6995b5826f45cb82b0c8", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "16fa232cfd1809c87fccd63ecd47df5f1067fbdfd825a370041a023969ef37ba", "generation_seconds_batch": 28.741010984987952, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4641343951225281, "target_z": 0.35156254179246854, "ce": 6.173524379730225, "genuineness": 4.400654315948486, "identity": 0.34002646803855896, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f0901d327c16fea849ba5fac948e1ee41ff567a650dc6995b5826f45cb82b0c8.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "f0d90d6d27d0660cab8f2c119acc605b71291feb79a22eb3cc6445b6b1ed2e78", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "1583fa3dc581104881050cbf7af4be66220ce92256331021137ac643d0898118", "generation_seconds_batch": 22.662760969018564, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1378541886806488, "target_z": -0.27148438767112076, "ce": 6.181467056274414, "genuineness": 3.8420867919921875, "identity": 0.27004241943359375, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f0d90d6d27d0660cab8f2c119acc605b71291feb79a22eb3cc6445b6b1ed2e78.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "f13d3c8c0c16ea9b423f6a30cf8c22064b6bced3cdbd0e7924d4b78405a82c15", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "db018703c5f218e12232f9fedc320e3149fa1d616d28acfb4b84c94aa681f4d6", "generation_seconds_batch": 18.949474229943007, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0018188953399658203, "target_z": -0.5312499822159709, "ce": 6.046441555023193, "genuineness": 4.532451152801514, "identity": 0.5777069926261902, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f13d3c8c0c16ea9b423f6a30cf8c22064b6bced3cdbd0e7924d4b78405a82c15.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "f14dcc93f114982ef38c5239097af08bb1b896ec58ae0434a70833ba231fa560", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "fe3bb8a2f3b9c62618330ee8a340303fa8e35f06414292db43badd1eba54aed4", "generation_seconds_batch": 23.22524118691217, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0018188953399658203, "target_z": -0.5312499822159709, "ce": 5.808161735534668, "genuineness": 4.2249250411987305, "identity": 0.6033243536949158, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f14dcc93f114982ef38c5239097af08bb1b896ec58ae0434a70833ba231fa560.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "f1f3e6379be6509084c4ddba600a16b4ccdaa2ec0013851cd9ad3616b7c2702f", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "1b3c33886262a3b769b754e2e24089e02dd5b99db1033ec70b338ddf3690be19", "generation_seconds_batch": 26.71569083002396, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.023298144340515137, "target_z": -0.49023437700070327, "ce": 6.24500846862793, "genuineness": 4.65797233581543, "identity": 0.701977014541626, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f1f3e6379be6509084c4ddba600a16b4ccdaa2ec0013851cd9ad3616b7c2702f.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "f1f41f95a990fa277ebf3d32fe347daa2ec6df689e8fdcdc3a0ac9cd1dcd7e56", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "991d94fdce426b5af864c21c6c71b43e3e84a9d0f8047ae8ce54846787d68e6b", "generation_seconds_batch": 26.04662565898616, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02432096004486084, "target_z": -0.4882812637826226, "ce": 6.181467056274414, "genuineness": 4.337893962860107, "identity": 0.744315505027771, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f1f41f95a990fa277ebf3d32fe347daa2ec6df689e8fdcdc3a0ac9cd1dcd7e56.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "f215940b68d36c19946aa7caf9d646b6fbbc1da2dcf11837cd045f033ea13acb", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "71bcf34d45bfbaaec1e173a81ad1d3dce0a92ceb2a747e9c0efb12b8eb12efd4", "generation_seconds_batch": 27.715750394039787, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07750770449638367, "target_z": -0.38671875044460075, "ce": 6.046441555023193, "genuineness": 4.3881025314331055, "identity": 0.6239648461341858, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f215940b68d36c19946aa7caf9d646b6fbbc1da2dcf11837cd045f033ea13acb.mp3" }, { "id": "heldout-fr-0", "language": "fr", "sentences": [ "Ton message est arrivé après mon départ.", "Je l’ai relu... puis j’ai rangé mon téléphone.", "Dis-moi quand nous pourrons parler." ], "text": "Ton message est arrivé après mon départ. Je l’ai relu... puis j’ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-0", "key": "f21bc0adf2df6aeda4aa5bc68657c60b57559ffed3b647118e970ebfd0ba6ad8", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Intense Malevolence Malice. 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": "6063317d7a0b6d57915305a9f3d06f40a9680ba72316e13a929e1f229eb81f37", "generation_seconds_batch": 19.520779552985914, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04477740824222565, "target_z": -0.44921874333098905, "ce": 6.316492557525635, "genuineness": 4.570107460021973, "identity": 0.7520643472671509, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi, quand nous pourrons parler ?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f21bc0adf2df6aeda4aa5bc68657c60b57559ffed3b647118e970ebfd0ba6ad8.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "f248891fdc1d3e07c0ef667d686c44e6ff650e68fbc01f0624cecd6c5f01b78c", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "5509b3077394e28470ac1c4d2c23fe2e43b7b348f68ad31d31d0c1ad2281f811", "generation_seconds_batch": 25.077688689925708, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2720995545387268, "target_z": -0.0151367412579119, "ce": 5.95112943649292, "genuineness": 4.15588903427124, "identity": 0.4780053496360779, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f248891fdc1d3e07c0ef667d686c44e6ff650e68fbc01f0624cecd6c5f01b78c.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "f2b67e8312145611efd95f27ade6d84a941b31e4fc8ffa9fdd976a3a02335457", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "bdb60a49490caa3860fd756ba7499194a9a5b81d7afa69cfa48358636e8fc53b", "generation_seconds_batch": 26.761937272036448, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00636368989944458, "target_z": -0.5468750017784029, "ce": 6.133810997009277, "genuineness": 4.268857479095459, "identity": 0.7580217719078064, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f2b67e8312145611efd95f27ade6d84a941b31e4fc8ffa9fdd976a3a02335457.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "f2ca02ccfa7bb8b841de215a56f9cbf58238f6bda4a63db328250f75886ac9ad", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "66f9f1137a2e7e9771fd1581bca92f6c7c172a82b2153d0fe634b8c8c8df75c4", "generation_seconds_batch": 26.803152821026742, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00022676587104797363, "target_z": -0.5351562655610255, "ce": 6.078212261199951, "genuineness": 4.457139015197754, "identity": 0.5733717679977417, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f2ca02ccfa7bb8b841de215a56f9cbf58238f6bda4a63db328250f75886ac9ad.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "f32162068f0a1c6d761e90ce954d6af878a0224a69f9f9914241fb538bafad85", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "5b1014ada06e8a104cdeafd0c2e04b610ebab63806a494e9ec6814999b985e7a", "generation_seconds_batch": 25.27061033807695, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0018188953399658203, "target_z": -0.5312499822159709, "ce": 6.125868320465088, "genuineness": 4.319066047668457, "identity": 0.6091296672821045, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f32162068f0a1c6d761e90ce954d6af878a0224a69f9f9914241fb538bafad85.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": "f37ddf17c680bbdf27637a61b4157adcd64d4994d4383dfefc76081c8964c2f3", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The keys are on the kitchen table.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "406fe486c8cf2d4facebae43011293b0291562fd0f4a103ff05bc42e1e838c3b", "generation_seconds_batch": 17.22400686191395, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.008409321308135986, "target_z": -0.5507812282145643, "ce": 6.3641486167907715, "genuineness": 4.18099308013916, "identity": 0.39618536829948425, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you leave.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f37ddf17c680bbdf27637a61b4157adcd64d4994d4383dfefc76081c8964c2f3.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "f3cfc3c9871dc7f0d648465674fed32cbfa427f526b0f6382ceac67b11114c90", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "c2840a5a784eeb7bebd4f35d0dc549d6d35b8525ab1af0851b7ec9b0096aa3ac", "generation_seconds_batch": 20.675008430029266, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1317172646522522, "target_z": -0.28320312388849816, "ce": 5.903473377227783, "genuineness": 4.068024158477783, "identity": 0.6952373385429382, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f3cfc3c9871dc7f0d648465674fed32cbfa427f526b0f6382ceac67b11114c90.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "f3fe721ef5f774bd34bbc6c6c3e548a96313173ac1cbf9d21ae5a4305e9997f9", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "2683aa84af30eb0b863965d24585412c6ab5d6e170311b8c2d6668d499e1fd5c", "generation_seconds_batch": 23.272989260964096, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.043185263872146606, "target_z": -0.6171874759915607, "ce": 6.252951145172119, "genuineness": 4.18099308013916, "identity": 0.5012052655220032, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f3fe721ef5f774bd34bbc6c6c3e548a96313173ac1cbf9d21ae5a4305e9997f9.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "f472cfe81e5410b071e9348b9f8f9676c96c7e462b6293c16285b6500f7b788a", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "2001e49418e645154d342ef43603121f30534d669b784e72878def00798c9406", "generation_seconds_batch": 20.826012525940314, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04682305455207825, "target_z": -0.44531248844038107, "ce": 6.133810997009277, "genuineness": 4.425758361816406, "identity": 0.7461339235305786, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f472cfe81e5410b071e9348b9f8f9676c96c7e462b6293c16285b6500f7b788a.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "f51433cbc8c67a4e6bd68f7e28931b3d4ad605038e8c625b715c51adf2ab0ad4", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "3a6162007d8d9e37974b00f5b0761708629123bd87fa7c135dd83de17be72e9a", "generation_seconds_batch": 25.02052228199318, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.016138404607772827, "target_z": -0.5039062264361613, "ce": 6.181467056274414, "genuineness": 4.149612903594971, "identity": 0.5755013227462769, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f51433cbc8c67a4e6bd68f7e28931b3d4ad605038e8c625b715c51adf2ab0ad4.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "f5b70ea693926bfbc231e5622976e9d5098a7e658b71e018d22e747f05afca61", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "a3cd14381818f60539061219746aa0b014f5ea622f408fa19a67256d39fa537d", "generation_seconds_batch": 29.204729752033018, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.036914587020874, "target_z": 1.4453124315314878, "ce": 6.324435234069824, "genuineness": 4.3504462242126465, "identity": 0.3002042770385742, "hypothesis": "Die Tür ist noch offen, ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f5b70ea693926bfbc231e5622976e9d5098a7e658b71e018d22e747f05afca61.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "f5c00495d4280a28941c247ff273aa4ddd07f44d2c38b1dd907d5b3bc84935ac", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "3ab46e42a9b1c67528804cd9a206ea0b3ef2807c569e58b09793e5fac29f70b5", "generation_seconds_batch": 30.35906647599768, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.007955819368362427, "target_z": -0.5195312459985935, "ce": 6.308549880981445, "genuineness": 4.482243061065674, "identity": 0.7137036323547363, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f5c00495d4280a28941c247ff273aa4ddd07f44d2c38b1dd907d5b3bc84935ac.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "f5e7e6441b24ce11c84611085d33a0fd3cff49bf4fe1d8b7cc35c1af214c37a2", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "4ab7e8e9378886dc573c6c9dbe5ef345e5207edf327d9a5156b5aebe0a0fe512", "generation_seconds_batch": 27.86458622198552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06830230355262756, "target_z": -0.4042968832251135, "ce": 6.1894097328186035, "genuineness": 4.287685871124268, "identity": 0.7421553134918213, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f5e7e6441b24ce11c84611085d33a0fd3cff49bf4fe1d8b7cc35c1af214c37a2.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "f60b266f0c7b39f5bf1baa7b697b2f1ffc765562186de329ed8438edbd2f3213", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "a4073ec949220f92a39247d7ebba10904824e956a9294245fa81f0d236a3dbf7", "generation_seconds_batch": 19.361784846987575, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.05955043435096741, "target_z": -0.6484375151164248, "ce": 6.459460258483887, "genuineness": 4.331617832183838, "identity": 0.28493309020996094, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f60b266f0c7b39f5bf1baa7b697b2f1ffc765562186de329ed8438edbd2f3213.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "f64c4c2fd497340ea90050f8146a9515b074d3d40af316d8566443072eda4dae", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "ccc6e94f4efaadc59d9157a28449ab436cec682184560d9aec6526a6ea207826", "generation_seconds_batch": 27.730090899043716, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.0329570472240448, "target_z": -0.5976562299929672, "ce": 6.09409761428833, "genuineness": 4.419482707977295, "identity": 0.6610955595970154, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f64c4c2fd497340ea90050f8146a9515b074d3d40af316d8566443072eda4dae.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": "f6927bb49ab21e1f4a528a2a8f709bfb67c8fa7802076bed21e0c16251d6ed9f", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "473f0e1ff61fa2312517b2b20255ac8fdd92502e616075e0e5f7e76632a01cc2", "generation_seconds_batch": 18.432049656985328, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03659482300281525, "target_z": -0.46484376289342116, "ce": 6.117925643920898, "genuineness": 4.231201171875, "identity": 0.40483781695365906, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier, neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f6927bb49ab21e1f4a528a2a8f709bfb67c8fa7802076bed21e0c16251d6ed9f.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": "f75523d84bee04c8ba5fe8916efc66f993979271b7af0d63951b3c9cece8a595", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3ff266e1e18ef6929bfe9dcad45993a0ae6bcdf78634aad64a6d00c8a0e5ec76", "generation_seconds_batch": 18.71437037095893, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01045498251914978, "target_z": -0.5546875115596189, "ce": 6.133810997009277, "genuineness": 3.9927117824554443, "identity": 0.5604346990585327, "hypothesis": "Las llaves están en la mesa de la cocina. Miré en el cajón y luego junto a la ventana. Llévatelas cuando te vayas.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f75523d84bee04c8ba5fe8916efc66f993979271b7af0d63951b3c9cece8a595.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": "f77f5b518e9756eb3950999418f42a7d09fc0aefdb931e8bf74d08ef8ee7ae5e", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c176267655469152494d919c1d21476246fa12fce4527ad9aebf80c5e2374225", "generation_seconds_batch": 17.272515240008943, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.22274841368198395, "target_z": -0.10937499466479125, "ce": 6.229123115539551, "genuineness": 4.507347106933594, "identity": 0.6375836133956909, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f77f5b518e9756eb3950999418f42a7d09fc0aefdb931e8bf74d08ef8ee7ae5e.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "f79c55c743c89d7fc7214945580346a068e8177ff5196b4381960411c23100ee", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "ffa637a99ddaa9b810870babc536c4c76df353ef4ecb77746000ecca0c80fab9", "generation_seconds_batch": 27.129553152946755, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4477692246437073, "target_z": 0.32031250266760436, "ce": 5.7128496170043945, "genuineness": 3.8797430992126465, "identity": 0.7352215647697449, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f79c55c743c89d7fc7214945580346a068e8177ff5196b4381960411c23100ee.mp3" }, { "id": "heldout-fr-1", "language": "fr", "sentences": [ "Les clés sont sur la table de la cuisine.", "J’ai regardé dans le tiroir... puis près de la fenêtre.", "Prends-les quand tu partiras." ], "text": "Les clés sont sur la table de la cuisine. J’ai regardé dans le tiroir... puis près de la fenêtre. Prends-les quand tu partiras.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-1", "key": "f7a65885d0e107603bdce9d92b5fc3b68a68524db7b0b06f79b6538794e8bf5c", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Intense Malevolence Malice. 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": "508d68abd0bf2e8519f8ccf8d6514816be6ae82772cde0d650eece81cde4f122", "generation_seconds_batch": 20.0645887440769, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.008409321308135986, "target_z": -0.5507812282145643, "ce": 6.109982967376709, "genuineness": 4.331617832183838, "identity": 0.7496307492256165, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai regardé dans le tiroir, puis près de la fenêtre. Prends-les quand tu partiras.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f7a65885d0e107603bdce9d92b5fc3b68a68524db7b0b06f79b6538794e8bf5c.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "f86b481a2e6160b7317fd89278cd77db17c94be48c28ff7f26b5aedaade20c9f", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "2e5baf0a0a1606973302199aa2ee781bf7ef5d91865f15779918f9766afc0f15", "generation_seconds_batch": 26.605277815950103, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.024774491786956787, "target_z": -0.5820312673394284, "ce": 6.173524379730225, "genuineness": 4.595211505889893, "identity": 0.5787200927734375, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f86b481a2e6160b7317fd89278cd77db17c94be48c28ff7f26b5aedaade20c9f.mp3" }, { "id": "heldout-de-2", "language": "de", "sentences": [ "Die Straße führt hinter der Brücke nach links.", "Geh weiter... bis zu dem kleinen Holztor.", "Ich warte auf der anderen Seite." ], "text": "Die Straße führt hinter der Brücke nach links. Geh weiter... bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-2", "key": "f86feddc7af334e5c3d6691e715ece3168f8e7618bd3b63ca7d0171f412fdf9d", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Intense Malevolence Malice. 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": "d5e79ffc06076dfcd4cb9932ca3f5b18e3dfd13fea024574f13021b7248314a1", "generation_seconds_batch": 20.43981575698126, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.014092743396759033, "target_z": -0.5078125097812161, "ce": 6.372091293334961, "genuineness": 4.432034492492676, "identity": 0.5283461809158325, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter, bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f86feddc7af334e5c3d6691e715ece3168f8e7618bd3b63ca7d0171f412fdf9d.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "f8e9331d3cbb823a1095dd7fb5ac9960ab611393ba7bcf7f28d2320ca51f33a7", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "1c07c3a68ab2a900e5e4821791139a1edcf67d8c9ce404ed57c6fffdbba57679", "generation_seconds_batch": 27.59941558202263, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06523384153842926, "target_z": -0.4101562513338022, "ce": 6.165581703186035, "genuineness": 4.15588903427124, "identity": 0.7524949312210083, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f8e9331d3cbb823a1095dd7fb5ac9960ab611393ba7bcf7f28d2320ca51f33a7.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "f9371fbc95af251062f61b5c737a168df09ca6b0824122b7afff4f70f640ed9c", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "3ca078a457d1886a5c08ea6e5c0ffc8d2eaa16bd9f790a7aa61c7a52263d3667", "generation_seconds_batch": 20.636983719072305, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03659482300281525, "target_z": -0.46484376289342116, "ce": 5.95112943649292, "genuineness": 4.18726921081543, "identity": 0.5578910112380981, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f9371fbc95af251062f61b5c737a168df09ca6b0824122b7afff4f70f640ed9c.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": "f9709cff8741a8d3c005b249d5c7d9e16bba7a6afb20af965b9d1449fdc9e7a9", "bucket": "emo_Malevolence_Malice", "phase": "heldout", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3052bf6d8919c2e42d090c64ca5e52e7aab08e2bfa09bb93d02330459a4905c2", "generation_seconds_batch": 26.381959442049265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08466745913028717, "target_z": -0.373046872554696, "ce": 6.332377910614014, "genuineness": 3.6851859092712402, "identity": 0.5307501554489136, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, lies sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f9709cff8741a8d3c005b249d5c7d9e16bba7a6afb20af965b9d1449fdc9e7a9.mp3" }, { "id": "emo_Malevolence_Malice-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-0", "key": "f9ac04a0e46ba1812086b8cfbd50a639ebde528431cd8f22b72b4a9842f6561f", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds 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": "a65d2d9823717b45044cc870377535486ff283271fda307cb38be7f096f5f175", "generation_seconds_batch": 29.262115109944716, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02125251293182373, "target_z": -0.49414060343686467, "ce": 6.062326908111572, "genuineness": 4.287685871124268, "identity": 0.659705400466919, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f9ac04a0e46ba1812086b8cfbd50a639ebde528431cd8f22b72b4a9842f6561f.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "f9e22cb75df33e2eb95808106a2fd911bdc8968f9adbb02a1f8b89fb1302e0f4", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "ecb97e802275841b988d96fed98ede5e8e1a619d3204d0332b4041d8ea1e7fe5", "generation_seconds_batch": 26.350344324018806, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9059934020042419, "target_z": 1.1953124599859344, "ce": 6.24500846862793, "genuineness": 3.9676074981689453, "identity": 0.38373488187789917, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/f9e22cb75df33e2eb95808106a2fd911bdc8968f9adbb02a1f8b89fb1302e0f4.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "fa11e336d0d4a984913648403fd7c7a8107b2b538b5cb636d08d80c51d5e42ef", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3723e3b293591f9d3d3d96981468aa4d0297c17c636914dcb25851bc5b17dee2", "generation_seconds_batch": 25.418444887036458, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.031097412109375, "target_z": 3.3437498612845724, "ce": 6.117925643920898, "genuineness": 3.9927117824554443, "identity": 0.1843334287405014, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/fa11e336d0d4a984913648403fd7c7a8107b2b538b5cb636d08d80c51d5e42ef.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "fa5904d0660baea4865d16c9a1f19e61368e7608d04930032fd0fec2f00aa6b5", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "31779227b42bfea851c7f6ebde405754b0f8b4c7b27b915e796e14ad4ca67c61", "generation_seconds_batch": 27.02010803401936, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.41094762086868286, "target_z": 0.24999997154555334, "ce": 6.165581703186035, "genuineness": 4.293961524963379, "identity": 0.6975783705711365, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/fa5904d0660baea4865d16c9a1f19e61368e7608d04930032fd0fec2f00aa6b5.mp3" }, { "id": "emo_Malevolence_Malice-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-1", "key": "fa7784f725a6cb0587a0e2380c7d8cdd1257ae9deadf72b4c0f4f50ae5aa2ecf", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] I found the little box.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] It was beside the old window.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [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": "9f3b5f7bbc90e7dd30a07cec03cd48779f6572ca10b210a51c0eb9a53e7ab1eb", "generation_seconds_batch": 25.36039623094257, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.0452309250831604, "target_z": -0.6210937593366153, "ce": 6.348263263702393, "genuineness": 4.193545341491699, "identity": 0.4016527533531189, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/fa7784f725a6cb0587a0e2380c7d8cdd1257ae9deadf72b4c0f4f50ae5aa2ecf.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "fa8681150abe68009f28973dacc81a363fe56dbbc8443cabda803d43409d9e29", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "5a75dc150dafa13fc1ef7e38980b07f92d6b13636c260fd47b15949af24c7be0", "generation_seconds_batch": 25.940985750989057, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03557200729846954, "target_z": -0.46679687611150184, "ce": 5.9431867599487305, "genuineness": 4.080576419830322, "identity": 0.5445735454559326, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/fa8681150abe68009f28973dacc81a363fe56dbbc8443cabda803d43409d9e29.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "fabfc65b2948d686337cf3f390cc96f5daebf00c65fde5dba914b34dad44ebf8", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": "17025d6458d064e5e9ba633a3d1243e42bc4b918ac3f556b3540ef2a4a8a8577", "generation_seconds_batch": 29.301259291009046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.014092743396759033, "target_z": -0.5078125097812161, "ce": 6.09409761428833, "genuineness": 4.017816066741943, "identity": 0.7867885828018188, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/fabfc65b2948d686337cf3f390cc96f5daebf00c65fde5dba914b34dad44ebf8.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "fb4728aa2d427e0683ec6f0238ced22f1fc840a9bd8aa99adb992b701b7ef7be", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "830e6c5bbe07b913719000aabaa433c153a588af6be0aff7f346f14c23d994cf", "generation_seconds_batch": 25.241973585099913, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.007955819368362427, "target_z": -0.5195312459985935, "ce": 6.213237762451172, "genuineness": 4.017816066741943, "identity": 0.25747764110565186, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/fb4728aa2d427e0683ec6f0238ced22f1fc840a9bd8aa99adb992b701b7ef7be.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "fcd775e4d09a5d00eeaab2788598506fe05799b6eb2b1aeb02e3b77a3c3d78b4", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": "7bd800efbed1381cb6c9af97d5c0300ab707cafcee5602e160a0959ea1650333", "generation_seconds_batch": 23.22524118691217, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.016591906547546387, "target_z": -0.5664062477769963, "ce": 5.863759994506836, "genuineness": 4.293961524963379, "identity": 0.6755621433258057, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/fcd775e4d09a5d00eeaab2788598506fe05799b6eb2b1aeb02e3b77a3c3d78b4.mp3" }, { "id": "emo_Malevolence_Malice-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-0", "key": "fd10d68ec8933942f9afea0d29ad40f71ae8c1538bd7e30201dd84610e767fe0", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds 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": "a6438238beedab71ff45e069c33486935aa41202c51c616acf446803dda40a82", "generation_seconds_batch": 26.83741046988871, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00636368989944458, "target_z": -0.5468750017784029, "ce": 6.316492557525635, "genuineness": 4.413206577301025, "identity": 0.5459094643592834, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/fd10d68ec8933942f9afea0d29ad40f71ae8c1538bd7e30201dd84610e767fe0.mp3" }, { "id": "emo_Malevolence_Malice-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-0", "key": "fd54aa3478f4a196e1770911e8f3b1bf2aff1232b0bad3259fd6af74908cbb24", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] De deur staat nog open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds 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": 127, "seconds_generated": 10.16, "mp3_sha256": "82b16340e952732b1552ed1b3a7ad1e531f2a884cf39110ca49ae959e96d9248", "generation_seconds_batch": 19.642417176044546, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.42117583751678467, "target_z": 0.2695312175441468, "ce": 6.09409761428833, "genuineness": 4.105680465698242, "identity": 0.5602501630783081, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/fd54aa3478f4a196e1770911e8f3b1bf2aff1232b0bad3259fd6af74908cbb24.mp3" }, { "id": "emo_Malevolence_Malice-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-en-0", "key": "fddc95fda7e454aee4d51ba849fbbab351e8812bb984ec3503eb4f862406d1c2", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] The door is still open.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] I put your coat on the chair.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 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": "18596e168d02f9acf52bc5a9ff1c4076c69c13cc9a364d28a6c0f85a453efb03", "generation_seconds_batch": 17.95443709997926, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.012047111988067627, "target_z": -0.5117187362173774, "ce": 6.2767791748046875, "genuineness": 4.017816066741943, "identity": 0.15942230820655823, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/fddc95fda7e454aee4d51ba849fbbab351e8812bb984ec3503eb4f862406d1c2.mp3" }, { "id": "emo_Malevolence_Malice-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-0", "key": "fde60a4d2ffccb6fbdd3e357ddf5e164679dbd95917135238b6e8dac135b6b3a", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [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": "4ecb6ebe8be58d5fb87628b65aacd1f0045b59ae7fb50b58285e45d566a209ba", "generation_seconds_batch": 26.091396669042297, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00022676587104797363, "target_z": -0.5351562655610255, "ce": 5.967014789581299, "genuineness": 4.030367851257324, "identity": 0.5583042502403259, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/fde60a4d2ffccb6fbdd3e357ddf5e164679dbd95917135238b6e8dac135b6b3a.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "fe26d65217e87bffd29a7d9f76b540f79462e7a6ec54d688e076c1455d73288d", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "5aa42d8b3242e4bd85b6e3717bece4350fdff780cc8567a2c7fe60d223c1c46e", "generation_seconds_batch": 17.171336216968484, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06727948784828186, "target_z": -0.40624999644319415, "ce": 6.0146708488464355, "genuineness": 4.005263805389404, "identity": 0.4362606406211853, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/fe26d65217e87bffd29a7d9f76b540f79462e7a6ec54d688e076c1455d73288d.mp3" }, { "id": "emo_Malevolence_Malice-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-de-1", "key": "fe57006322d943554163d19b59d4786fe64d8a149a5a151867709ac2084da7d8", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear 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": "6a78630788b6e62055772ff8460d142b323873ce7fdf257d30ec2e1abef6e614", "generation_seconds_batch": 25.041195454075933, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4416322708129883, "target_z": 0.3085937095413337, "ce": 6.205295085906982, "genuineness": 4.080576419830322, "identity": 0.23093776404857635, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/fe57006322d943554163d19b59d4786fe64d8a149a5a151867709ac2084da7d8.mp3" }, { "id": "emo_Malevolence_Malice-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-es-1", "key": "fea66b8242be633981006015829ac98e8fd5f813426a28bedbbddc3ad7392cb4", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 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": 127, "seconds_generated": 10.16, "mp3_sha256": "7968577cf91fc61266b3f41cb51402a0fad0cc0b07f4cd43347560d2606b1c90", "generation_seconds_batch": 27.208889418980107, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.018637537956237793, "target_z": -0.5703124742131577, "ce": 6.173524379730225, "genuineness": 4.344170093536377, "identity": 0.6444287300109863, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/fea66b8242be633981006015829ac98e8fd5f813426a28bedbbddc3ad7392cb4.mp3" }, { "id": "emo_Malevolence_Malice-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-fr-1", "key": "fee95a0bd5bdc7ccd89025838f053e3c56fa2fbca8a8722b94d0cba1ec7995b0", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] 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": "1f6f75dc0fcfc3432836a16bfe44e866dc75528634dc4adac77f342b894af59d", "generation_seconds_batch": 26.189175114035606, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00636368989944458, "target_z": -0.5468750017784029, "ce": 6.24500846862793, "genuineness": 4.482243061065674, "identity": 0.6808476448059082, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/fee95a0bd5bdc7ccd89025838f053e3c56fa2fbca8a8722b94d0cba1ec7995b0.mp3" }, { "id": "emo_Malevolence_Malice-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "emo_Malevolence_Malice-nl-1", "key": "ffcb98569976db63443153ce82d2527be11b9af52c69d34ca31b21792d0e4aa2", "bucket": "emo_Malevolence_Malice", "phase": "grid", "candidate": "manual-11c936b4d8e6", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Focus on precise articulation, making each consonant sharp and biting, suggesting intellectual cruelty over raw aggression.\nSCRIPT:\n(Deliver the opening with unnaturally crisp diction, each syllable sounding carved from ice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce subtle vocal tension across the mid-range frequencies, suggesting restrained, powerful enjoyment.) [3.8 seconds duration] Het stond naast het oude raam.\n(End with a perfectly controlled, almost surgical delivery, leaving no room for emotional bleed.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 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": "72c427139db368a99e02db68bbbea99ca040561532a47255b3694574a77d227e", "generation_seconds_batch": 26.260819707997143, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02432096004486084, "target_z": -0.4882812637826226, "ce": 6.125868320465088, "genuineness": 4.124508857727051, "identity": 0.6312105655670166, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Malevolence_Malice/ffcb98569976db63443153ce82d2527be11b9af52c69d34ca31b21792d0e4aa2.mp3" }, { "key": "original-emo_Malevolence_Malice-de-0-seed1701-scale0.0", "bucket": "emo_Malevolence_Malice", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "a0813541ae9eaa1c58503b18caa0a09d46ed603f16c264baf62e58f03d66916b", "task_id": "emo_Malevolence_Malice-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.057051271200180054, "target_z": -0.4257812424417876, "ce": 6.046441555023193, "genuineness": 4.8588056564331055, "identity": 0.6335868835449219, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/de/21ae309fbd07020749b33d2f.mp3" }, { "key": "original-emo_Malevolence_Malice-de-0-seed1701-scale0.5", "bucket": "emo_Malevolence_Malice", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "1ef3da34dec54426e77a0cd172747ccbb50ef834586e258bed92b2319ab424f1", "task_id": "emo_Malevolence_Malice-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.012047111988067627, "target_z": -0.5117187362173774, "ce": 6.348263263702393, "genuineness": 4.394378662109375, "identity": 0.5946799516677856, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/de/3cc68f6b5f51f452f2f9a15d.mp3" }, { "key": "original-emo_Malevolence_Malice-de-0-seed1701-scale0.75", "bucket": "emo_Malevolence_Malice", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "fe8913208a61a71d19b3a4bd2f2dc338f963b93b28d810c0d11ce41d5c456133", "task_id": "emo_Malevolence_Malice-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.059096917510032654, "target_z": -0.4218749875511796, "ce": 6.252951145172119, "genuineness": 4.482243061065674, "identity": 0.6206172108650208, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/de/d4de7f0b6036286c634d68e9.mp3" }, { "key": "original-emo_Malevolence_Malice-de-0-seed1701-scale1.0", "bucket": "emo_Malevolence_Malice", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "0d2b8253dd6fea164b90bf5f7bac6a8183e948b0b1f7db1ddad0cb7bbbf7a67e", "task_id": "emo_Malevolence_Malice-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02022966742515564, "target_z": -0.49609377356383866, "ce": 6.316492557525635, "genuineness": 4.281409740447998, "identity": 0.5399235486984253, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/de/2e6bcb5583d7fe5a0e7533be.mp3" }, { "key": "original-emo_Malevolence_Malice-de-0-seed1702-scale0.0", "bucket": "emo_Malevolence_Malice", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "7099c7aeb733833580ca1387ec4888181ee1df16f8cb0a154953a806d5a7b739", "task_id": "emo_Malevolence_Malice-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.27069318294525146, "target_z": -0.017822271932772842, "ce": 6.157639026641846, "genuineness": 4.325342178344727, "identity": 0.42359602451324463, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/de/3c9d4344e5c6814dd4b6dd97.mp3" }, { "key": "original-emo_Malevolence_Malice-de-0-seed1702-scale0.5", "bucket": "emo_Malevolence_Malice", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "c2f38802104193b16017e8f7c72d6ccd1beee5a91e827ff5d48a60dbfdc13ca9", "task_id": "emo_Malevolence_Malice-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5664165616035461, "target_z": 0.5468750017784029, "ce": 6.133810997009277, "genuineness": 4.406930446624756, "identity": 0.48218870162963867, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/de/3981a6b58998d3d80bcfe00e.mp3" }, { "key": "original-emo_Malevolence_Malice-de-0-seed1702-scale0.75", "bucket": "emo_Malevolence_Malice", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "c95ca3c2938675872b7a00ddd12bd2b6c89b3de730e0fca98f376930354bdc4e", "task_id": "emo_Malevolence_Malice-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04273176193237305, "target_z": -0.4531249982215971, "ce": 6.205295085906982, "genuineness": 4.519899368286133, "identity": 0.5574551820755005, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/de/95c25c64138311ccad9b6ad9.mp3" }, { "key": "original-emo_Malevolence_Malice-de-0-seed1702-scale1.0", "bucket": "emo_Malevolence_Malice", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "3e6ff38a90c8be20238e1d7da1df5e79fc6ce9571009698152abbe773eba2312", "task_id": "emo_Malevolence_Malice-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8078024983406067, "target_z": 1.0078124528723227, "ce": 6.23706579208374, "genuineness": 4.369274139404297, "identity": 0.4822891354560852, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/de/d55784ac7bec50ac369f68f9.mp3" }, { "key": "original-emo_Malevolence_Malice-de-1-seed1701-scale0.0", "bucket": "emo_Malevolence_Malice", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "3e14316b9d7a5556e8861d92e22c069ddd57e5937d39a4f09b6eda085856a012", "task_id": "emo_Malevolence_Malice-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.041708946228027344, "target_z": -0.45507811143967775, "ce": 6.467403411865234, "genuineness": 4.293961524963379, "identity": 0.6154637336730957, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/de/8761a0c60c31e2ef2c8be00c.mp3" }, { "key": "original-emo_Malevolence_Malice-de-1-seed1701-scale0.5", "bucket": "emo_Malevolence_Malice", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "b5af0f375a7ad80050d1ecfddc603f09b9659a302038ffe64ffb22024b04d0af", "task_id": "emo_Malevolence_Malice-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04886868596076965, "target_z": -0.4414062620042197, "ce": 6.475345611572266, "genuineness": 4.425758361816406, "identity": 0.6352746486663818, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/de/0b68a8c22197aa1cc3cc3c37.mp3" }, { "key": "original-emo_Malevolence_Malice-de-1-seed1701-scale0.75", "bucket": "emo_Malevolence_Malice", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "1f210103562c1ec91e2d488a4f8f4030841927d20e29001f66edb859d8189e1f", "task_id": "emo_Malevolence_Malice-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03659482300281525, "target_z": -0.46484376289342116, "ce": 6.475345611572266, "genuineness": 4.18099308013916, "identity": 0.5629994869232178, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/de/b0fd2d49bba9777575551b2e.mp3" }, { "key": "original-emo_Malevolence_Malice-de-1-seed1701-scale1.0", "bucket": "emo_Malevolence_Malice", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "4c588388ecf382399f0feed8000798d1b36e4b51012a4db2e899052aca1b921e", "task_id": "emo_Malevolence_Malice-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06318819522857666, "target_z": -0.4140625062244102, "ce": 6.467403411865234, "genuineness": 4.482243061065674, "identity": 0.5356155633926392, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/de/77d7231bc107e9b09ee48aee.mp3" }, { "key": "original-emo_Malevolence_Malice-de-1-seed1702-scale0.0", "bucket": "emo_Malevolence_Malice", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "e5275746782de538fbf31586957d0ff4d5ea295cd265b4b01a59e886d970b2ab", "task_id": "emo_Malevolence_Malice-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09898696839809418, "target_z": -0.3457031167748865, "ce": 6.419747352600098, "genuineness": 4.093128204345703, "identity": 0.4673413634300232, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/de/188a42d976f10b2d0eea5526.mp3" }, { "key": "original-emo_Malevolence_Malice-de-1-seed1702-scale0.5", "bucket": "emo_Malevolence_Malice", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "4063982c00459b960432bc2f51210445178c28cc7295e55c7224495b1c1211a6", "task_id": "emo_Malevolence_Malice-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07750770449638367, "target_z": -0.38671875044460075, "ce": 6.38003396987915, "genuineness": 4.532451152801514, "identity": 0.5233230590820312, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/de/0b915009cfed1d9176e1f598.mp3" }, { "key": "original-emo_Malevolence_Malice-de-1-seed1702-scale0.75", "bucket": "emo_Malevolence_Malice", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "e69e7c034a3aa5552f91fd26be36c0207c8b343bdc6b11f57b8ce536b83f3964", "task_id": "emo_Malevolence_Malice-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.059096917510032654, "target_z": -0.4218749875511796, "ce": 6.41180419921875, "genuineness": 4.3504462242126465, "identity": 0.4859897494316101, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/de/cb5ab43f6154b1a265ec6ac6.mp3" }, { "key": "original-emo_Malevolence_Malice-de-1-seed1702-scale1.0", "bucket": "emo_Malevolence_Malice", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "de287c8c4e63574d1007fa5abb9fb3b2dfdc53e14c8863173acde891a66aa76c", "task_id": "emo_Malevolence_Malice-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08466745913028717, "target_z": -0.373046872554696, "ce": 6.443574905395508, "genuineness": 4.482243061065674, "identity": 0.40058594942092896, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/de/b15b8ac26ded503fdee588ad.mp3" }, { "key": "original-emo_Malevolence_Malice-en-0-seed1701-scale0.0", "bucket": "emo_Malevolence_Malice", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "f903a43cc3be80222b040b935cc8e9a49f3cff5e885172636b120bacdf47bc4d", "task_id": "emo_Malevolence_Malice-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.026820123195648193, "target_z": -0.5859374937755898, "ce": 6.181467056274414, "genuineness": 3.8671910762786865, "identity": 0.44523221254348755, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/en/e5f614eacabc36d888f8918d.mp3" }, { "key": "original-emo_Malevolence_Malice-en-0-seed1701-scale0.5", "bucket": "emo_Malevolence_Malice", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "10927f7930994a5415a49bb2fe50421bc3ce4bd3aa7130919784b4cede24ee14", "task_id": "emo_Malevolence_Malice-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.018637537956237793, "target_z": -0.5703124742131577, "ce": 6.300607204437256, "genuineness": 3.9048471450805664, "identity": 0.31987330317497253, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/en/27375ef33a939f478354e2e2.mp3" }, { "key": "original-emo_Malevolence_Malice-en-0-seed1701-scale0.75", "bucket": "emo_Malevolence_Malice", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "f6db52e5d4aeaee082dbe69964a7b7adb29a5e60c9d58f636f9a10ab9d05fdba", "task_id": "emo_Malevolence_Malice-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.035002708435058594, "target_z": -0.6015625133380219, "ce": 6.348263263702393, "genuineness": 4.193545341491699, "identity": 0.23609575629234314, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/en/7811054888bb98f88b1a5d9b.mp3" }, { "key": "original-emo_Malevolence_Malice-en-0-seed1701-scale1.0", "bucket": "emo_Malevolence_Malice", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "ed160b1322913a120ac18ca80125d434c22e39c250bc68e65d29a323042042c2", "task_id": "emo_Malevolence_Malice-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.016138404607772827, "target_z": -0.5039062264361613, "ce": 6.1894097328186035, "genuineness": 4.030367851257324, "identity": 0.4274313449859619, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/en/29320170ad7365d4de5af6d2.mp3" }, { "key": "original-emo_Malevolence_Malice-en-0-seed1702-scale0.0", "bucket": "emo_Malevolence_Malice", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "0e0047326154f3a9dcd09ecfab411b326842ffc0ec1f500a4404d7c70145b466", "task_id": "emo_Malevolence_Malice-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03352636098861694, "target_z": -0.47070313100210986, "ce": 6.205295085906982, "genuineness": 4.293961524963379, "identity": 0.31140029430389404, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/en/b495b34577577ec67f4f6759.mp3" }, { "key": "original-emo_Malevolence_Malice-en-0-seed1702-scale0.5", "bucket": "emo_Malevolence_Malice", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "de4ceefc2e0e279b153d6471179f09152a0f7976cd3dceae17fc45502f3672d1", "task_id": "emo_Malevolence_Malice-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.018184036016464233, "target_z": -0.5, "ce": 6.229123115539551, "genuineness": 3.7793264389038086, "identity": 0.4886592626571655, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/en/4475ce24befc7e49df805e98.mp3" }, { "key": "original-emo_Malevolence_Malice-en-0-seed1702-scale0.75", "bucket": "emo_Malevolence_Malice", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "dcca3b64351f5eef6df1b6ed36fcbe17fff8152a19ac4c4cbcf421ab6ae8f867", "task_id": "emo_Malevolence_Malice-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00022676587104797363, "target_z": -0.5351562655610255, "ce": 6.117925643920898, "genuineness": 3.9425034523010254, "identity": 0.3392521142959595, "hypothesis": "The door is still open. I've put your coat on the chair. Tell me when you are ready.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/en/7ce106e459e1bc17455ccd9c.mp3" }, { "key": "original-emo_Malevolence_Malice-en-0-seed1702-scale1.0", "bucket": "emo_Malevolence_Malice", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "6726414d4f3052cff50d9d33a0128f3fb23e2328a703ef43b5735160dc37b828", "task_id": "emo_Malevolence_Malice-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.018184036016464233, "target_z": -0.5, "ce": 6.221180438995361, "genuineness": 4.005263805389404, "identity": 0.27113041281700134, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/en/2a25d44b345d9808ca57f274.mp3" }, { "key": "original-emo_Malevolence_Malice-en-1-seed1701-scale0.0", "bucket": "emo_Malevolence_Malice", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "764fa56a90a076406335315d870d49bf212acf42884d1ea7a13b3f9cca6a94a4", "task_id": "emo_Malevolence_Malice-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.06568735837936401, "target_z": -0.6601562513338022, "ce": 6.427689552307129, "genuineness": 4.582659721374512, "identity": 0.5747895836830139, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/en/f95e489b80155d9646ab6560.mp3" }, { "key": "original-emo_Malevolence_Malice-en-1-seed1701-scale0.5", "bucket": "emo_Malevolence_Malice", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "ca4076b360b17b8aad6ab179ae1797c382fdb0548b1020d450769ae0b0ade75d", "task_id": "emo_Malevolence_Malice-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.08000686764717102, "target_z": -0.6875000071136117, "ce": 6.300607204437256, "genuineness": 4.293961524963379, "identity": 0.3485945463180542, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/en/467bec6f178341a5760c4161.mp3" }, { "key": "original-emo_Malevolence_Malice-en-1-seed1701-scale0.75", "bucket": "emo_Malevolence_Malice", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "43c3ed1c132123b07a5fc32eccbeb4a6c65bed3739f8dbf61cf4629745eaaa8b", "task_id": "emo_Malevolence_Malice-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.039094001054763794, "target_z": -0.6093750231192379, "ce": 6.403861999511719, "genuineness": 4.519899368286133, "identity": 0.47930628061294556, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/en/b95fe5f0248e52a58f7f4da6.mp3" }, { "key": "original-emo_Malevolence_Malice-en-1-seed1701-scale1.0", "bucket": "emo_Malevolence_Malice", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "e57add6fb6571fabee8242d571541b4872dd921ab0e1cc2609819c22697f1fab", "task_id": "emo_Malevolence_Malice-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.0452309250831604, "target_z": -0.6210937593366153, "ce": 6.292664527893066, "genuineness": 4.124508857727051, "identity": 0.4720369577407837, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/en/a4cdcb38bf87fa5540fc92b5.mp3" }, { "key": "original-emo_Malevolence_Malice-en-1-seed1702-scale0.0", "bucket": "emo_Malevolence_Malice", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "2b6850c809a920722098483cba372217105926c549db0b66d2419d984c836409", "task_id": "emo_Malevolence_Malice-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.061596065759658813, "target_z": -0.6523437415525861, "ce": 6.3641486167907715, "genuineness": 4.406930446624756, "identity": 0.26789921522140503, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/en/c143af05bb8a07714bef0522.mp3" }, { "key": "original-emo_Malevolence_Malice-en-1-seed1702-scale0.5", "bucket": "emo_Malevolence_Malice", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "2ad48932c2b840a83f81266892ef59bc3401a9eabf5175f5efd922bac3e4923c", "task_id": "emo_Malevolence_Malice-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.035002708435058594, "target_z": -0.6015625133380219, "ce": 6.356205940246582, "genuineness": 4.15588903427124, "identity": 0.14480963349342346, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/en/d0d8b02140fb1477820594b2.mp3" }, { "key": "original-emo_Malevolence_Malice-en-1-seed1702-scale0.75", "bucket": "emo_Malevolence_Malice", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "1dafaa2a87c3207bdd3084d35d2cc15b7061f9bd84380a0327b3c7ccebe9d9b0", "task_id": "emo_Malevolence_Malice-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.07591557502746582, "target_z": -0.6796874973323956, "ce": 6.372091293334961, "genuineness": 4.444586753845215, "identity": 0.5063887238502502, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/en/6eebbe1c56f973d1cbeb51b2.mp3" }, { "key": "original-emo_Malevolence_Malice-en-1-seed1702-scale1.0", "bucket": "emo_Malevolence_Malice", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "7f9a0ac4122590380031d13525e022ada90d62d6bd864fb435160022256aa9b9", "task_id": "emo_Malevolence_Malice-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.04727655649185181, "target_z": -0.6249999857727767, "ce": 6.356205940246582, "genuineness": 4.218649387359619, "identity": 0.38191112875938416, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/en/893c4f962a5773b41c24ff17.mp3" }, { "key": "original-emo_Malevolence_Malice-es-0-seed1701-scale0.0", "bucket": "emo_Malevolence_Malice", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "32bc96714966ddbe054c3ad4e588cb40437553f26fccdc5b5b1383498faaa667", "task_id": "emo_Malevolence_Malice-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09080438315868378, "target_z": -0.3613281363373186, "ce": 6.125868320465088, "genuineness": 4.049196243286133, "identity": 0.5948885679244995, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/es/5dbeb357b3dde9fcfd90589f.mp3" }, { "key": "original-emo_Malevolence_Malice-es-0-seed1701-scale0.5", "bucket": "emo_Malevolence_Malice", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "af67af8788b131a50ca498d1a4edbb41e855b1734a154f6035de3c31c6e4c798", "task_id": "emo_Malevolence_Malice-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.026820123195648193, "target_z": -0.5859374937755898, "ce": 6.0146708488464355, "genuineness": 4.137060642242432, "identity": 0.5349474549293518, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/es/3adcbfb26b8935083383d28f.mp3" }, { "key": "original-emo_Malevolence_Malice-es-0-seed1701-scale0.75", "bucket": "emo_Malevolence_Malice", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "e3536f9389db14e86d716f291f8ee3fafe06d7ccd8e1db47d4921c620541ed59", "task_id": "emo_Malevolence_Malice-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.018637537956237793, "target_z": -0.5703124742131577, "ce": 5.95112943649292, "genuineness": 4.2249250411987305, "identity": 0.5414972305297852, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/es/666b35d958774ae43053c0ff.mp3" }, { "key": "original-emo_Malevolence_Malice-es-0-seed1701-scale1.0", "bucket": "emo_Malevolence_Malice", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "99e9b7bcd412d992c711bad03e85433846ce327cd65a67d7f2e11f6a7adeb1f2", "task_id": "emo_Malevolence_Malice-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0018188953399658203, "target_z": -0.5312499822159709, "ce": 6.046441555023193, "genuineness": 4.093128204345703, "identity": 0.5696797966957092, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/es/a11e6d37447b23c757c04cb9.mp3" }, { "key": "original-emo_Malevolence_Malice-es-0-seed1702-scale0.0", "bucket": "emo_Malevolence_Malice", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "8c42edf676b8ab32150571a40f1ecd9ee4041b15acdc9a11390285c48022e7dd", "task_id": "emo_Malevolence_Malice-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02841225266456604, "target_z": -0.48046875400140654, "ce": 5.982900142669678, "genuineness": 4.030367851257324, "identity": 0.5962163805961609, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/es/ed70b7a68b53f9e4da012420.mp3" }, { "key": "original-emo_Malevolence_Malice-es-0-seed1702-scale0.5", "bucket": "emo_Malevolence_Malice", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "1dc3a7f5c0ae6c6afc96ade913421dc94b38904e48f5fb8f38477c9add52d0be", "task_id": "emo_Malevolence_Malice-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.018637537956237793, "target_z": -0.5703124742131577, "ce": 6.078212261199951, "genuineness": 4.168440818786621, "identity": 0.4497184455394745, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/es/6248ed3e6da340a8026e482e.mp3" }, { "key": "original-emo_Malevolence_Malice-es-0-seed1702-scale0.75", "bucket": "emo_Malevolence_Malice", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "61160b2afb80f158e8036688fedc6a3195e2e894bad95644827f7b9a7cdacf55", "task_id": "emo_Malevolence_Malice-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00636368989944458, "target_z": -0.5468750017784029, "ce": 5.998785495758057, "genuineness": 3.998987913131714, "identity": 0.41404658555984497, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/es/359c896d83f22de57c3049fd.mp3" }, { "key": "original-emo_Malevolence_Malice-es-0-seed1702-scale1.0", "bucket": "emo_Malevolence_Malice", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "808eacd3557347c0fd969b713888f150f9e36ea445970509aa6885760be805f7", "task_id": "emo_Malevolence_Malice-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.016138404607772827, "target_z": -0.5039062264361613, "ce": 6.062326908111572, "genuineness": 4.118232727050781, "identity": 0.17020003497600555, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/es/b5755e352128dfebe2fbd7e1.mp3" }, { "key": "original-emo_Malevolence_Malice-es-1-seed1701-scale0.0", "bucket": "emo_Malevolence_Malice", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "27ef55627a9660f8daebd8364e1737688cfa6b7ca11c065bc20d4bb711ea2d36", "task_id": "emo_Malevolence_Malice-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.035002708435058594, "target_z": -0.6015625133380219, "ce": 6.213237762451172, "genuineness": 4.519899368286133, "identity": 0.5952341556549072, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/es/14e34f9683cfd67168a55101.mp3" }, { "key": "original-emo_Malevolence_Malice-es-1-seed1701-scale0.5", "bucket": "emo_Malevolence_Malice", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "91556db8369f47d6f77f3407c89bcf24af2881310c8d8f081449a2decf31cdc3", "task_id": "emo_Malevolence_Malice-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.04727655649185181, "target_z": -0.6249999857727767, "ce": 6.0305562019348145, "genuineness": 4.582659721374512, "identity": 0.7664927244186401, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/es/af91277e5de18afa4f2a3a2a.mp3" }, { "key": "original-emo_Malevolence_Malice-es-1-seed1701-scale0.75", "bucket": "emo_Malevolence_Malice", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "1bd61b55b600377935b2b6748912e50db21f8760d104c7cdb1c7bb64aab5d279", "task_id": "emo_Malevolence_Malice-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01045498251914978, "target_z": -0.5546875115596189, "ce": 6.117925643920898, "genuineness": 4.457139015197754, "identity": 0.6972050666809082, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/es/ef887a0b9e7b5487815e06b2.mp3" }, { "key": "original-emo_Malevolence_Malice-es-1-seed1701-scale1.0", "bucket": "emo_Malevolence_Malice", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "a688e1dcaebb67a1602ae399543e6a877e5d78a0e77ef231f130caaf83119cc8", "task_id": "emo_Malevolence_Malice-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01045498251914978, "target_z": -0.5546875115596189, "ce": 6.260893821716309, "genuineness": 4.607763767242432, "identity": 0.7604098916053772, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/es/293c9dfdd0a6d11653c1e2ff.mp3" }, { "key": "original-emo_Malevolence_Malice-es-1-seed1702-scale0.0", "bucket": "emo_Malevolence_Malice", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "05a430d74989fac29dad1798148c16bfa224d6a133c2c39b52173c7e066a482f", "task_id": "emo_Malevolence_Malice-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.016591906547546387, "target_z": -0.5664062477769963, "ce": 6.1020402908325195, "genuineness": 4.6956281661987305, "identity": 0.31333303451538086, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/es/7bf12098418821c522ed9e68.mp3" }, { "key": "original-emo_Malevolence_Malice-es-1-seed1702-scale0.5", "bucket": "emo_Malevolence_Malice", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "e373d2ecd16542b1db6d493db8523b3d22de6ba6b0404add4741f96e74e41d17", "task_id": "emo_Malevolence_Malice-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.004318058490753174, "target_z": -0.5429687753422415, "ce": 6.125868320465088, "genuineness": 4.362998008728027, "identity": 0.3866082429885864, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/es/20bddccd4e56092019b02faa.mp3" }, { "key": "original-emo_Malevolence_Malice-es-1-seed1702-scale0.75", "bucket": "emo_Malevolence_Malice", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "1f344ff084c4fe11b29e60e82f6fd47be3a489254d9f575e02edfdc53c79e148", "task_id": "emo_Malevolence_Malice-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.026820123195648193, "target_z": -0.5859374937755898, "ce": 6.252951145172119, "genuineness": 4.482243061065674, "identity": 0.7346659898757935, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/es/fd09a1ade21ce2c49ce7df3d.mp3" }, { "key": "original-emo_Malevolence_Malice-es-1-seed1702-scale1.0", "bucket": "emo_Malevolence_Malice", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "dcbe0d61543da9a5f7c007d1776036ec9f4f5850facc8ea9c7dc4ab53261abc9", "task_id": "emo_Malevolence_Malice-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.016591906547546387, "target_z": -0.5664062477769963, "ce": 6.356205940246582, "genuineness": 4.68307638168335, "identity": 0.729770839214325, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/es/2377f571b72f8afaada75219.mp3" }, { "key": "original-emo_Malevolence_Malice-fr-0-seed1701-scale0.0", "bucket": "emo_Malevolence_Malice", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "71862821bbbfc19127e1b3b2bd102f5fed494986329eb43f6dc0e82f41df51fd", "task_id": "emo_Malevolence_Malice-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.023298144340515137, "target_z": -0.49023437700070327, "ce": 6.213237762451172, "genuineness": 4.369274139404297, "identity": 0.8217543363571167, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/fr/79023befc7017e7987079a31.mp3" }, { "key": "original-emo_Malevolence_Malice-fr-0-seed1701-scale0.5", "bucket": "emo_Malevolence_Malice", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "d0985618428e3d6d71ed3e90d9109bf055793768276bdc51bb974499e6fdd5a0", "task_id": "emo_Malevolence_Malice-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00227239727973938, "target_z": -0.5390624919971869, "ce": 6.292664527893066, "genuineness": 4.7458367347717285, "identity": 0.5601322650909424, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/fr/fa0f3160a8504f27c813068d.mp3" }, { "key": "original-emo_Malevolence_Malice-fr-0-seed1701-scale0.75", "bucket": "emo_Malevolence_Malice", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "37cf2ffb1a7b1dea1b0192af0844b251ed298c5bf12bd9fcf6f2efca47d1fd47", "task_id": "emo_Malevolence_Malice-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0018188953399658203, "target_z": -0.5312499822159709, "ce": 6.2767791748046875, "genuineness": 4.325342178344727, "identity": 0.6040369272232056, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/fr/623b04984aaee594ab7093cc.mp3" }, { "key": "original-emo_Malevolence_Malice-fr-0-seed1701-scale1.0", "bucket": "emo_Malevolence_Malice", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "3e827c242837a8fa598422aa2a93c22fd69deba0c4e10bed97592399a400b5e3", "task_id": "emo_Malevolence_Malice-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00636368989944458, "target_z": -0.5468750017784029, "ce": 6.165581703186035, "genuineness": 4.243753433227539, "identity": 0.6239906549453735, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/fr/7fe79d151fd135d036015117.mp3" }, { "key": "original-emo_Malevolence_Malice-fr-0-seed1702-scale0.0", "bucket": "emo_Malevolence_Malice", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "86f64d1bb38c4381b3f1e3e32557c65adea4594d6c358c0f94016619e5a915d8", "task_id": "emo_Malevolence_Malice-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.012500613927841187, "target_z": -0.5585937379957803, "ce": 6.1894097328186035, "genuineness": 4.356721878051758, "identity": 0.6427960991859436, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/fr/d60009a4dfc0ff1ec5c636b6.mp3" }, { "key": "original-emo_Malevolence_Malice-fr-0-seed1702-scale0.5", "bucket": "emo_Malevolence_Malice", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "4c2603f9d41880e226103f2c4183d0b90b66b6c3842ee1ba3432d4e012be3823", "task_id": "emo_Malevolence_Malice-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.012047111988067627, "target_z": -0.5117187362173774, "ce": 6.165581703186035, "genuineness": 4.293961524963379, "identity": 0.6556476354598999, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/fr/a6275685e0b788aa4710a89e.mp3" }, { "key": "original-emo_Malevolence_Malice-fr-0-seed1702-scale0.75", "bucket": "emo_Malevolence_Malice", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "f9205dbf0c99f774fca0b7918e5b3168b4d967e48deb635e5d984dab13a77944", "task_id": "emo_Malevolence_Malice-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.010001450777053833, "target_z": -0.5156250195624321, "ce": 6.09409761428833, "genuineness": 4.118232727050781, "identity": 0.778697669506073, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/fr/faa25787c79305203e07a6fc.mp3" }, { "key": "original-emo_Malevolence_Malice-fr-0-seed1702-scale1.0", "bucket": "emo_Malevolence_Malice", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "7977a58328f08db3571c7e8d1efdc164e16e3c60567761f1f2362ab8600892a5", "task_id": "emo_Malevolence_Malice-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.039663299918174744, "target_z": -0.4589843663302858, "ce": 6.0146708488464355, "genuineness": 4.306513786315918, "identity": 0.5945684909820557, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/fr/a690037312cf02175129496f.mp3" }, { "key": "original-emo_Malevolence_Malice-fr-1-seed1701-scale0.0", "bucket": "emo_Malevolence_Malice", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "2291b08bb11b89f59f138b252bd97cb449c74d915ddf7a76bab529a2c0f14304", "task_id": "emo_Malevolence_Malice-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.004318058490753174, "target_z": -0.5429687753422415, "ce": 6.284721851348877, "genuineness": 4.532451152801514, "identity": 0.6433247327804565, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/fr/70a376a04b1b9a3f0cef3196.mp3" }, { "key": "original-emo_Malevolence_Malice-fr-1-seed1701-scale0.5", "bucket": "emo_Malevolence_Malice", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "2c6cded0f57fca09c78309a3c47b61f87e38d6eea2edba9412bcdbebf963ad0c", "task_id": "emo_Malevolence_Malice-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.0411396324634552, "target_z": -0.6132812495553993, "ce": 6.348263263702393, "genuineness": 4.607763767242432, "identity": 0.6302741765975952, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/fr/909118f362a5a9129767946b.mp3" }, { "key": "original-emo_Malevolence_Malice-fr-1-seed1701-scale0.75", "bucket": "emo_Malevolence_Malice", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "5664f85a467f032bade66d89cc2f70c9b8aa11975d07f505d324133f82dc92e4", "task_id": "emo_Malevolence_Malice-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.012500613927841187, "target_z": -0.5585937379957803, "ce": 6.181467056274414, "genuineness": 4.582659721374512, "identity": 0.757232666015625, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/fr/2fcbd4561dcd75d11fd950fe.mp3" }, { "key": "original-emo_Malevolence_Malice-fr-1-seed1701-scale1.0", "bucket": "emo_Malevolence_Malice", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "31391869f9c3a461897be4ad066688035f0cc9dd9fddd1c255dc31b3f1d672c3", "task_id": "emo_Malevolence_Malice-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01454627513885498, "target_z": -0.562500021340835, "ce": 6.24500846862793, "genuineness": 4.444586753845215, "identity": 0.731925368309021, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/fr/08f6420192c5b3de5608a01c.mp3" }, { "key": "original-emo_Malevolence_Malice-fr-1-seed1702-scale0.0", "bucket": "emo_Malevolence_Malice", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "5ad7929555c6f3be116934deb9e8fedde182850fbc5c480af981d6219586dc0d", "task_id": "emo_Malevolence_Malice-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.016591906547546387, "target_z": -0.5664062477769963, "ce": 6.157639026641846, "genuineness": 4.582659721374512, "identity": 0.8068645000457764, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/fr/4c77a98e92b1816d6e7d4e61.mp3" }, { "key": "original-emo_Malevolence_Malice-fr-1-seed1702-scale0.5", "bucket": "emo_Malevolence_Malice", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "d190a8cc3f053b8bcbc7f8a9776e0a98a41ef15f066ad89730eb88365613e832", "task_id": "emo_Malevolence_Malice-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.018637537956237793, "target_z": -0.5703124742131577, "ce": 6.221180438995361, "genuineness": 4.444586753845215, "identity": 0.6998276710510254, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/fr/aef63e3ca0f2f2352b1ce34d.mp3" }, { "key": "original-emo_Malevolence_Malice-fr-1-seed1702-scale0.75", "bucket": "emo_Malevolence_Malice", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "e5129ca9ea531e5547d1e9314b12b10eb006ea870b65da3c6d4e8f19d57c0cb9", "task_id": "emo_Malevolence_Malice-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0018188953399658203, "target_z": -0.5312499822159709, "ce": 6.213237762451172, "genuineness": 4.519899368286133, "identity": 0.7142552137374878, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/fr/09d780fdc6093c2ddd4dd8b7.mp3" }, { "key": "original-emo_Malevolence_Malice-fr-1-seed1702-scale1.0", "bucket": "emo_Malevolence_Malice", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "4e2454ddb1fc6777f5c9aaba5cb6611223ed2f64c5d8054030fbd858830df9da", "task_id": "emo_Malevolence_Malice-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01454627513885498, "target_z": -0.562500021340835, "ce": 6.221180438995361, "genuineness": 4.319066047668457, "identity": 0.6897094249725342, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/fr/f0293e9b1b897243f8a90013.mp3" }, { "key": "original-emo_Malevolence_Malice-nl-0-seed1701-scale0.0", "bucket": "emo_Malevolence_Malice", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "6d84dcfad07d932ca1129a1487d0ff1d4b7992d95639deff6f3537da9fde112d", "task_id": "emo_Malevolence_Malice-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.01045498251914978, "target_z": -0.5546875115596189, "ce": 6.252951145172119, "genuineness": 4.331617832183838, "identity": 0.36898085474967957, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/nl/f37bb5ca742adb32691109b0.mp3" }, { "key": "original-emo_Malevolence_Malice-nl-0-seed1701-scale0.5", "bucket": "emo_Malevolence_Malice", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "38f6209406e45d8f1ae2c3672998f46dca8966820d89134ceff4da5baaf1aa50", "task_id": "emo_Malevolence_Malice-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.026820123195648193, "target_z": -0.5859374937755898, "ce": 6.157639026641846, "genuineness": 4.306513786315918, "identity": 0.5417598485946655, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/nl/f8607a4c1b0053bff2b6851f.mp3" }, { "key": "original-emo_Malevolence_Malice-nl-0-seed1701-scale0.75", "bucket": "emo_Malevolence_Malice", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "268de521435b4d82ef3fa09a0dab037508dda10b5bc89dfc350fec4dd697ab74", "task_id": "emo_Malevolence_Malice-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04682305455207825, "target_z": -0.44531248844038107, "ce": 6.133810997009277, "genuineness": 4.444586753845215, "identity": 0.7651753425598145, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/nl/cb9b298b9bfd11a07de238c8.mp3" }, { "key": "original-emo_Malevolence_Malice-nl-0-seed1701-scale1.0", "bucket": "emo_Malevolence_Malice", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "258f76f0986a7483c3b969c7ca4c6bf3b0eb4bc92c9cdb56debc33768595a4f7", "task_id": "emo_Malevolence_Malice-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03864046931266785, "target_z": -0.46093750800281313, "ce": 6.157639026641846, "genuineness": 4.168440818786621, "identity": 0.3286708891391754, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/nl/9b710fc07e3b1e74f2b20be3.mp3" }, { "key": "original-emo_Malevolence_Malice-nl-0-seed1702-scale0.0", "bucket": "emo_Malevolence_Malice", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "edf9ca8978e13ba90b968076856a90472f2c0817ce284d2ea558a9693ea771f3", "task_id": "emo_Malevolence_Malice-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.029435068368911743, "target_z": -0.47851564078332587, "ce": 6.229123115539551, "genuineness": 4.319066047668457, "identity": 0.7302052974700928, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/nl/e327e07623d27988f3d24853.mp3" }, { "key": "original-emo_Malevolence_Malice-nl-0-seed1702-scale0.5", "bucket": "emo_Malevolence_Malice", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "4e0918fbcddcbe6cf5940a2a8c43a54a235a1e2a0f6fb6541be0b756f1f1d7bb", "task_id": "emo_Malevolence_Malice-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00022676587104797363, "target_z": -0.5351562655610255, "ce": 6.109982967376709, "genuineness": 4.293961524963379, "identity": 0.6689068078994751, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/nl/ad627964ff6d46141dc421a6.mp3" }, { "key": "original-emo_Malevolence_Malice-nl-0-seed1702-scale0.75", "bucket": "emo_Malevolence_Malice", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "80364f37173b9c0f7e5648159e3a896deae830e11bd9d77784620e14c1a86846", "task_id": "emo_Malevolence_Malice-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.026366621255874634, "target_z": -0.48437498043756794, "ce": 5.935244083404541, "genuineness": 4.18099308013916, "identity": 0.599311113357544, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/nl/bb7c9ecaa7ca0975d86a167e.mp3" }, { "key": "original-emo_Malevolence_Malice-nl-0-seed1702-scale1.0", "bucket": "emo_Malevolence_Malice", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "3ead58e3fff49a72eadbd1a34ede11e696955aa2d74861d32216390d709e174d", "task_id": "emo_Malevolence_Malice-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.016138404607772827, "target_z": -0.5039062264361613, "ce": 5.998785495758057, "genuineness": 4.20609712600708, "identity": 0.6921392679214478, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/nl/febd4fe0b71c34d004c7570a.mp3" }, { "key": "original-emo_Malevolence_Malice-nl-1-seed1701-scale0.0", "bucket": "emo_Malevolence_Malice", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "1bdd3222fddb9218708795c9d0390eb246b973e009c22beb8b00dd10796705d1", "task_id": "emo_Malevolence_Malice-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.019206851720809937, "target_z": -0.49804688678191933, "ce": 6.133810997009277, "genuineness": 4.1998209953308105, "identity": 0.661110520362854, "hypothesis": "ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/nl/77349107ff3290b628596363.mp3" }, { "key": "original-emo_Malevolence_Malice-nl-1-seed1701-scale0.5", "bucket": "emo_Malevolence_Malice", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "b39a307134150ce354e0cce76b31b1589e535ff4150716e583e2d56ea9b4d396", "task_id": "emo_Malevolence_Malice-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.03557200729846954, "target_z": -0.46679687611150184, "ce": 6.109982967376709, "genuineness": 4.369274139404297, "identity": 0.6853176951408386, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/nl/1a092f2b41e8d4e749907a69.mp3" }, { "key": "original-emo_Malevolence_Malice-nl-1-seed1701-scale0.75", "bucket": "emo_Malevolence_Malice", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "3b0b226d855e564f0d467bb2b11174f968551eb15d7e9ad97c180eaf4d37930a", "task_id": "emo_Malevolence_Malice-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04068611562252045, "target_z": -0.4570312531122051, "ce": 6.1894097328186035, "genuineness": 4.18099308013916, "identity": 0.6175937652587891, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/nl/be98002a7dd4ee1d95710451.mp3" }, { "key": "original-emo_Malevolence_Malice-nl-1-seed1701-scale1.0", "bucket": "emo_Malevolence_Malice", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "b31a24f0712d483256cd31e5f75d91e7748d7f1f4072d6c5108af791cdeee880", "task_id": "emo_Malevolence_Malice-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04886868596076965, "target_z": -0.4414062620042197, "ce": 6.229123115539551, "genuineness": 4.444586753845215, "identity": 0.6973481178283691, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/nl/466001cf542c147753e4c634.mp3" }, { "key": "original-emo_Malevolence_Malice-nl-1-seed1702-scale0.0", "bucket": "emo_Malevolence_Malice", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "9c55ceccaa3a32e8e5b80dfeb1904bcd5b310e6bfbd68f849529807a9a4ec36c", "task_id": "emo_Malevolence_Malice-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.00022676587104797363, "target_z": -0.5351562655610255, "ce": 6.23706579208374, "genuineness": 4.394378662109375, "identity": 0.5899051427841187, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/nl/a8c1b004b0f307c6838e5706.mp3" }, { "key": "original-emo_Malevolence_Malice-nl-1-seed1702-scale0.5", "bucket": "emo_Malevolence_Malice", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "b65196e48fda0ac2096e2a45cbea715010645140a974cd07fbd67057395b27ed", "task_id": "emo_Malevolence_Malice-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.019206851720809937, "target_z": -0.49804688678191933, "ce": 6.109982967376709, "genuineness": 4.306513786315918, "identity": 0.6451966166496277, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/nl/468ff67678633f5af56236ab.mp3" }, { "key": "original-emo_Malevolence_Malice-nl-1-seed1702-scale0.75", "bucket": "emo_Malevolence_Malice", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "0fb5bb8bd00b0dc7e688b1379706f1b483854b26da2d65d87778de11a1e86ad9", "task_id": "emo_Malevolence_Malice-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.023298144340515137, "target_z": -0.49023437700070327, "ce": 6.125868320465088, "genuineness": 4.369274139404297, "identity": 0.40578514337539673, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/nl/b7c78f008840233b3267bd37.mp3" }, { "key": "original-emo_Malevolence_Malice-nl-1-seed1702-scale1.0", "bucket": "emo_Malevolence_Malice", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "937105358f3e7099dee7bd61bd073a5a97c4836f0e6d72197685158fba816d2a", "mp3_sha256": "dd99d93e19b65d2f8bd21c8af2eda4022ee47f46f41e8b2e70cef61172e800a0", "task_id": "emo_Malevolence_Malice-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Intense Malevolence Malice. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0059101879596710205, "target_z": -0.5234374724347548, "ce": 6.157639026641846, "genuineness": 4.068024158477783, "identity": 0.6852218508720398, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Malevolence_Malice/nl/610cf1ea5931e30ce5fefa9a.mp3" } ], "source_standard_audio_unchanged": true }