{ "complete": true, "report": { "bucket": "vn_STNC_reg_low", "label": "submissive and deferential", "definition": { "id": "vn_STNC_reg_low", "category": "bipolar", "head": "vn_STNC_reg", "pole": "low", "label": "submissive and deferential", "floor": 1.0, "canonical_dimension_name": "Stance" }, "recipe_status": "new_canonical_dimension_extension", "manual_provenance": [], "selected_wording": { "id": "canonical-extension-P04", "general": "Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.", "inline": [ "Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.", "Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.", "Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis." ], "recipe": "P04", "cfg": 1.0, "timing_syntax": "canonical", "canonical_labels": true, "lexical_proxy": false, "origin": "new_canonical_manual_extension_requires_new_evidence" }, "selection": { "manual_no_adapter": { "config": { "scale": 0.0, "cfg": 5.0 }, "n": 20, "target_z": 1.2722656326884092, "target_raw": 1.12532639503479, "ce": 5.608998990058899, "genuineness": 4.64573380947113, "identity": 0.566514540463686, "wer": 0.01112831097351221, "valid_fraction": 1, "identity_delta": -0.01110774502158165, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "manual_adapter": { "config": { "scale": 0.5, "cfg": 5.0 }, "n": 20, "target_z": 1.3457031176387573, "target_raw": 1.0417065620422363, "ce": 5.560151600837708, "genuineness": 4.572304153442383, "identity": 0.5979676231741905, "wer": 0.02234907120743034, "valid_fraction": 1, "identity_delta": 0.020345337688922882, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "grid": [ { "config": { "scale": 0.0, "cfg": 5.0 }, "n": 20, "target_z": 1.2722656326884092, "target_raw": 1.12532639503479, "ce": 5.608998990058899, "genuineness": 4.64573380947113, "identity": 0.566514540463686, "wer": 0.01112831097351221, "valid_fraction": 1, "identity_delta": -0.01110774502158165, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 4.0 }, "n": 20, "target_z": 1.2166015666713708, "target_raw": 1.1887084662914276, "ce": 5.672143268585205, "genuineness": 4.595839238166809, "identity": 0.5712709635496139, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.0063513219356536865, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 3.0 }, "n": 20, "target_z": 1.0002929787694692, "target_raw": 1.4350089490413667, "ce": 5.865150141716003, "genuineness": 4.570735144615173, "identity": 0.583117938041687, "wer": 0.013742690058479532, "valid_fraction": 1, "identity_delta": 0.0054956525564193726, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.5 }, "n": 20, "target_z": 0.8990234529586097, "target_raw": 1.5503198266029359, "ce": 5.87686550617218, "genuineness": 4.586111426353455, "identity": 0.6177039325237275, "wer": 0.007894736842105263, "valid_fraction": 1, "identity_delta": 0.040081647038459775, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.0 }, "n": 20, "target_z": 0.6606933560419929, "target_raw": 1.8216951727867126, "ce": 5.981907391548157, "genuineness": 4.545944738388061, "identity": 0.6299359619617462, "wer": 0.002631578947368421, "valid_fraction": 1, "identity_delta": 0.052313676476478575, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.5 }, "n": 20, "target_z": 0.47849121560984437, "target_raw": 2.029160237312317, "ce": 6.04743447303772, "genuineness": 4.537785887718201, "identity": 0.5999752081930637, "wer": 0.002631578947368421, "valid_fraction": 1, "identity_delta": 0.022352922707796097, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.25 }, "n": 20, "target_z": 0.2385742048863262, "target_raw": 2.302342528104782, "ce": 6.0954876184463505, "genuineness": 4.545003366470337, "identity": 0.6134243488311768, "wer": 0.00804093567251462, "valid_fraction": 1, "identity_delta": 0.035802063345909116, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.0 }, "n": 20, "target_z": 0.23817138197995, "target_raw": 2.3028012037277223, "ce": 6.13341383934021, "genuineness": 4.548455142974854, "identity": 0.5751906272023917, "wer": 0.02660818713450292, "valid_fraction": 0.95, "identity_delta": -0.002431658282876015, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 5.0 }, "n": 20, "target_z": 1.3457031176387573, "target_raw": 1.0417065620422363, "ce": 5.560151600837708, "genuineness": 4.572304153442383, "identity": 0.5979676231741905, "wer": 0.02234907120743034, "valid_fraction": 1, "identity_delta": 0.020345337688922882, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 5.0 }, "n": 20, "target_z": 1.3312499973826692, "target_raw": 1.0581636548042297, "ce": 5.601751327514648, "genuineness": 4.638516354560852, "identity": 0.6308711290359497, "wer": 0.011721491228070174, "valid_fraction": 1, "identity_delta": 0.053248843550682066, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 5.0 }, "n": 20, "target_z": 1.3033203143812064, "target_raw": 1.0899658799171448, "ce": 5.575689482688904, "genuineness": 4.564459061622619, "identity": 0.6027756303548812, "wer": 0.019499269005847954, "valid_fraction": 0.95, "identity_delta": 0.025153344869613646, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 4.0 }, "n": 20, "target_z": 1.2212890647083727, "target_raw": 1.1833710312843322, "ce": 5.649705219268799, "genuineness": 4.586111354827881, "identity": 0.6033956691622734, "wer": 0.01477124183006536, "valid_fraction": 1, "identity_delta": 0.025773383677005768, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 4.0 }, "n": 20, "target_z": 1.2201171915077877, "target_raw": 1.18470538854599, "ce": 5.708481025695801, "genuineness": 4.6385164022445675, "identity": 0.5943964630365371, "wer": 0.008680555555555556, "valid_fraction": 1, "identity_delta": 0.01677417755126953, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 4.0 }, "n": 20, "target_z": 1.2158203143812065, "target_raw": 1.1895980417728425, "ce": 5.698552680015564, "genuineness": 4.643537163734436, "identity": 0.6018690124154091, "wer": 0.010982112143102855, "valid_fraction": 1, "identity_delta": 0.02424672693014145, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 3.0 }, "n": 20, "target_z": 1.0988281425033994, "target_raw": 1.3228115618228913, "ce": 5.813919878005981, "genuineness": 4.593642663955689, "identity": 0.5736907951533794, "wer": 0.005555555555555555, "valid_fraction": 1, "identity_delta": -0.0039314903318881985, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 3.0 }, "n": 20, "target_z": 1.0365234424892869, "target_raw": 1.3937550127506255, "ce": 5.8355636358261105, "genuineness": 4.615608787536621, "identity": 0.6070812821388245, "wer": 0.019298245614035085, "valid_fraction": 0.95, "identity_delta": 0.029458996653556824, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 3.0 }, "n": 20, "target_z": 1.027636709466655, "target_raw": 1.4038739204406738, "ce": 5.784134864807129, "genuineness": 4.5547312021255495, "identity": 0.5916068434715271, "wer": 0.03528874269005848, "valid_fraction": 0.9, "identity_delta": 0.013984557986259461, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.5 }, "n": 20, "target_z": 0.9857421797297994, "target_raw": 1.451577264070511, "ce": 5.880638360977173, "genuineness": 4.569166088104248, "identity": 0.6004418209195137, "wer": 0.005409356725146199, "valid_fraction": 1, "identity_delta": 0.022819535434246065, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.5 }, "n": 20, "target_z": 0.915136730405301, "target_raw": 1.5319723904132843, "ce": 5.944179677963257, "genuineness": 4.58171820640564, "identity": 0.6447761088609696, "wer": 0.003125, "valid_fraction": 1, "identity_delta": 0.0671538233757019, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.5 }, "n": 20, "target_z": 0.9145263793437745, "target_raw": 1.5326673686504364, "ce": 5.883219695091247, "genuineness": 4.592387437820435, "identity": 0.5867366850376129, "wer": 0.019399509803921567, "valid_fraction": 0.95, "identity_delta": 0.009114399552345276, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.0 }, "n": 20, "target_z": 0.7391113230334738, "target_raw": 1.7324042975902558, "ce": 6.006331110000611, "genuineness": 4.573873162269592, "identity": 0.6069966182112694, "wer": 0.013157894736842105, "valid_fraction": 0.95, "identity_delta": 0.02937433272600174, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.0 }, "n": 20, "target_z": 0.6729003720378621, "target_raw": 1.8077956140041351, "ce": 5.950335264205933, "genuineness": 4.553162217140198, "identity": 0.6380954399704933, "wer": 0.005409356725146199, "valid_fraction": 1, "identity_delta": 0.06047315448522568, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.0 }, "n": 20, "target_z": 0.6310058580049909, "target_raw": 1.8554989397525787, "ce": 5.93861985206604, "genuineness": 4.496991610527038, "identity": 0.6117562994360923, "wer": 0.022076023391812867, "valid_fraction": 0.95, "identity_delta": 0.034134013950824736, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.5 }, "n": 20, "target_z": 0.484912119455813, "target_raw": 2.021849054098129, "ce": 6.091119122505188, "genuineness": 4.552534604072571, "identity": 0.6052733436226845, "wer": 0.002631578947368421, "valid_fraction": 1, "identity_delta": 0.02765105813741684, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.5 }, "n": 20, "target_z": 0.48312988472437834, "target_raw": 2.023878401517868, "ce": 6.101444602012634, "genuineness": 4.561634874343872, "identity": 0.6062962114810944, "wer": 0.008742690058479531, "valid_fraction": 1, "identity_delta": 0.028673925995826723, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.5 }, "n": 20, "target_z": 0.4079223513345665, "target_raw": 2.109513705968857, "ce": 6.031747674942016, "genuineness": 4.535275506973266, "identity": 0.5684033215045929, "wer": 0.003125, "valid_fraction": 1, "identity_delta": -0.009218963980674743, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.25 }, "n": 20, "target_z": 0.36909177890129896, "target_raw": 2.153728264570236, "ce": 6.105415964126587, "genuineness": 4.548141407966614, "identity": 0.6054356031119823, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.027813317626714705, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.25 }, "n": 20, "target_z": 0.3651855540931075, "target_raw": 2.158176100254059, "ce": 6.0784108400344845, "genuineness": 4.529313230514527, "identity": 0.6343046948313713, "wer": 0.013859649122807016, "valid_fraction": 1, "identity_delta": 0.05668240934610367, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.0 }, "n": 20, "target_z": 0.28869018928117207, "target_raw": 2.245277798175812, "ce": 6.097274708747864, "genuineness": 4.520840644836426, "identity": 0.5552763029932976, "wer": 0.005263157894736842, "valid_fraction": 1, "identity_delta": -0.022345982491970062, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.25 }, "n": 20, "target_z": 0.27557374507366006, "target_raw": 2.2602128803730013, "ce": 6.078212285041809, "genuineness": 4.517388820648193, "identity": 0.5873716242611409, "wer": 0.013888888888888888, "valid_fraction": 0.95, "identity_delta": 0.009749338775873185, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.0 }, "n": 20, "target_z": 0.19377440660924172, "target_raw": 2.35335396528244, "ce": 6.120705604553223, "genuineness": 4.513309288024902, "identity": 0.5519444420933723, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.025677843391895293, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.0 }, "n": 20, "target_z": 0.11733396386575982, "target_raw": 2.440393126010895, "ce": 6.097274732589722, "genuineness": 4.473770332336426, "identity": 0.6063809156417846, "wer": 0.005409356725146199, "valid_fraction": 1, "identity_delta": 0.02875863015651703, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" } ], "development_standard": [ { "scale": 0.0, "n": 20, "target_z": 0.008374024102056642, "target_raw": 2.5644607186317443, "ce": 6.169155907630921, "genuineness": 4.465611433982849, "identity": 0.5776222854852676, "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": -0.07690427039497405, "target_raw": 2.661563128232956, "ce": 6.171935796737671, "genuineness": 4.44709712266922, "identity": 0.5627563774585724, "wer": 0.0029411764705882353, "valid_fraction": 1, "identity_delta": -0.01486590802669525, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 0.75, "n": 20, "target_z": 0.02830810191592798, "target_raw": 2.5417627155780793, "ce": 6.158830380439758, "genuineness": 4.487577557563782, "identity": 0.5774335734546184, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.00018871203064918518, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 1.0, "n": 20, "target_z": -0.05700684568092268, "target_raw": 2.6389068603515624, "ce": 6.153270554542542, "genuineness": 4.446783328056336, "identity": 0.565081462264061, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.012540823221206665, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" } ] }, "heldout": { "standard": { "n": 50, "target_z": -0.25194213060773585, "target_raw": 2.8608705615997314, "ce": 6.205930471420288, "genuineness": 4.440946626663208, "identity": 0.59952592253685, "wer": 0.006666666666666666, "valid_fraction": 1, "identity_delta": 0.0, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "manual": { "n": 50, "target_z": 0.9128515525377848, "target_raw": 1.5345744156837464, "ce": 5.731196994781494, "genuineness": 4.596090307235718, "identity": 0.6139840245246887, "wer": 0.024463579898362505, "valid_fraction": 0.98, "identity_delta": 0.014458101987838746, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "adapter": { "n": 50, "target_z": 0.9827343694054556, "target_raw": 1.4550021171569825, "ce": 5.713762836456299, "genuineness": 4.586174192428589, "identity": 0.6285103705525398, "wer": 0.02752656344418358, "valid_fraction": 0.98, "identity_delta": 0.028984448015689848, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "matched_no_adapter": { "n": 50, "target_z": 0.9128515525377848, "target_raw": 1.5345744156837464, "ce": 5.731196994781494, "genuineness": 4.596090307235718, "identity": 0.6139840245246887, "wer": 0.024463579898362505, "valid_fraction": 0.98, "identity_delta": 0.014458101987838746, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" } }, "adapter_vs_matched_cfg": { "n": 50, "delta_z": 0.06988281686767069, "ci95_z": [ -0.015312511385388777, 0.16484376210515472 ], "language_deltas": { "en": 0.11523437254625242, "de": 0.03671875294449709, "fr": 0.03437500654332688, "nl": -0.07617189413923113, "es": 0.23925784644350817 }, "languages_improved": 4, "clear_gain": false, "delta_ce": -0.017434158325195313, "delta_genuineness": -0.009916114807128906, "delta_wer": 0.0030629835458210746, "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": 0.06988281686767069, "ci95_z": [ -0.015312511385388777, 0.16484376210515472 ], "language_deltas": { "en": 0.11523437254625242, "de": 0.03671875294449709, "fr": 0.03437500654332688, "nl": -0.07617189413923113, "es": 0.23925784644350817 }, "languages_improved": 4, "clear_gain": false, "delta_ce": -0.017434158325195313, "delta_genuineness": -0.009916114807128906, "delta_wer": 0.0030629835458210746, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "manual_vs_standard": { "n": 50, "delta_z": 1.1647936831455206, "ci95_z": [ 1.0087646387308837, 1.3277294927948025 ], "language_deltas": { "en": 0.9003600923645553, "de": 1.0889892454926424, "fr": 1.2225097458927798, "nl": 1.442382824196197, "es": 1.169726507781429 }, "languages_improved": 5, "clear_gain": true, "delta_ce": -0.47473347663879395, "delta_genuineness": 0.15514368057250977, "delta_wer": 0.01779691323169584, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "route": { "adapter": false, "scale": 0.0, "cfg": 5.0, "reason": "Use manual-only configuration; no clear qualifying incremental adapter gain" }, "reference_only": true, "updated": 1793035529.5906525, "example": { "text": "Ton message est arrivé après mon départ. Je l’ai relu... puis j’ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "sentences": [ "Ton message est arrivé après mon départ.", "Je l’ai relu... puis j’ai rangé mon téléphone.", "Dis-moi quand nous pourrons parler." ], "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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.", "language": "fr", "seed": 1702, "scale": 0.0, "cfg": 5.0, "frames": 128, "reference": true } }, "rows": [ { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "001d8acae5271dc6c8782c087fdbbc10e338de31603250f5f1ae0a028085f949", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e61afd7964f0e7157f84085ce1a92b4561718da98588ad5b03dae2696334c8e2", "generation_seconds_batch": 23.757514734053984, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1239920854568481, "target_z": 1.273437464011702, "ce": 5.4825119972229, "genuineness": 4.821148872375488, "identity": 0.753943145275116, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/001d8acae5271dc6c8782c087fdbbc10e338de31603250f5f1ae0a028085f949.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "002165dd486d9e6114769723ee872090b6d74690f745e1bd0313068d2d67757f", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5414250173500ad1b0144b94ce12b75e3de55d79293e2a395aacd7aa791ded47", "generation_seconds_batch": 21.284522460075095, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1114180088043213, "target_z": 0.40624993456673125, "ce": 6.252951145172119, "genuineness": 4.6705241203308105, "identity": 0.7426769733428955, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/002165dd486d9e6114769723ee872090b6d74690f745e1bd0313068d2d67757f.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "0085ead92b1e4169d7f958873c5715a03ed9208e28fdaa0a269216a5c53663cd", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0aba8866c8ad1e0a5332498733d3e25409dfa1a4cf3682c1422ccd86957a3131", "generation_seconds_batch": 15.342324733966962, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3285938501358032, "target_z": 1.0937499607400387, "ce": 5.863759994506836, "genuineness": 4.68307638168335, "identity": 0.539219081401825, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/0085ead92b1e4169d7f958873c5715a03ed9208e28fdaa0a269216a5c53663cd.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "008cdca302a2ce7ea8553881307411e567469f4277d3522d704a44a877c22d2d", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "786b3e0a00be3d042210be2dcefad7419d0d8cd29fce3724ba4ab034b37447fe", "generation_seconds_batch": 17.313607475021854, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5350770950317383, "target_z": 0.03417961756819403, "ce": 6.046441555023193, "genuineness": 4.419482707977295, "identity": 0.8215956687927246, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/008cdca302a2ce7ea8553881307411e567469f4277d3522d704a44a877c22d2d.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "00fd6ea114504f95dbe670d1a11589cdebdaa245bf53ec7cb3412a393ed6780c", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "33d441a97c25e9db9ba7f8ed52f9833cda3cda9067b3d0682573c71c7ac58594", "generation_seconds_batch": 17.800279836985283, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2574279308319092, "target_z": 1.1562500392599613, "ce": 5.677107334136963, "genuineness": 4.833701133728027, "identity": 0.7120687365531921, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/00fd6ea114504f95dbe670d1a11589cdebdaa245bf53ec7cb3412a393ed6780c.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "01377a6a7d4469fb693207dcda87f30c50d9ea4b700087096b12668ad000b920", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0b877762d5105281d4712170a1e41efec65276f2f31de4255c8eabb84abe2e68", "generation_seconds_batch": 22.33053092902992, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9512948989868164, "target_z": 0.5468749280234044, "ce": 5.959072113037109, "genuineness": 4.833701133728027, "identity": 0.64194655418396, "hypothesis": "He encontrado la cajita. estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/01377a6a7d4469fb693207dcda87f30c50d9ea4b700087096b12668ad000b920.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1cce0a94b510295ba32aefec847360c56509efefe619357252d758e509b7e7c0", "generation_seconds_batch": 16.20624463201966, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.272310495376587, "target_z": -0.6132812156475339, "ce": 5.982900142669678, "genuineness": 4.6203155517578125, "identity": 0.664421796798706, "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/vn_STNC_reg_low/01f5d5f22d9b51f4e674eec92be80d0710e62f5704093e3409bc811236eec3e5.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "025c31ccdd1a81459e512a4430f6b19eab136a84ceca7302bf62c725ff7bbdf0", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "55f4df9ad85361a8e92f01020bec3ac3a3379f526838e0be783e3b34f3d18970", "generation_seconds_batch": 15.832757271942683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.527015209197998, "target_z": 0.04125981132604864, "ce": 6.372091293334961, "genuineness": 4.356721878051758, "identity": 0.4790712594985962, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/025c31ccdd1a81459e512a4430f6b19eab136a84ceca7302bf62c725ff7bbdf0.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": "026c348c1793c32c8d1fe18b5f5dc347c4b1f29c57dd8d3b46e1944218fddd18", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d20ac2d21c4baed490ebc9dcc8ba7a8795e62cf066e9a9b2b397068d37d3b6ce", "generation_seconds_batch": 16.28377297706902, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9549732208251953, "target_z": 1.4218749803700195, "ce": 5.331601142883301, "genuineness": 4.7332844734191895, "identity": 0.7281189560890198, "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/vn_STNC_reg_low/026c348c1793c32c8d1fe18b5f5dc347c4b1f29c57dd8d3b46e1944218fddd18.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "034359b0d814f7ca7c175d23c3b650a2516de21033ebbc646b80506470683042", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ad712fa154e69f114ada9260cac116227cc7301fd5512444087d068286f70408", "generation_seconds_batch": 22.393209049012512, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.670736789703369, "target_z": -0.08496085938903544, "ce": 5.760505676269531, "genuineness": 4.394378662109375, "identity": 0.6844691634178162, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/034359b0d814f7ca7c175d23c3b650a2516de21033ebbc646b80506470683042.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "037d0b1c24b6305139a444f7971b130b3f34fe81383214d54d48b0adc6a6e002", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "87c8b34a6d883c3cc6c9a29be8b7b87b0a39003589dd03971e224c9d8fc642c2", "generation_seconds_batch": 23.169558982015587, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5039420127868652, "target_z": 0.06152339537733324, "ce": 6.054384231567383, "genuineness": 4.281409740447998, "identity": 0.20939834415912628, "hypothesis": "The 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/vn_STNC_reg_low/037d0b1c24b6305139a444f7971b130b3f34fe81383214d54d48b0adc6a6e002.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "038e9db747aa5cf08c60964dd9666e704d8b783a4c7f8456b5cb79fdae93559b", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7a6d22afb773ed9480dfb0a0e09c817d081b38ce9f19b28298463cc3d46326af", "generation_seconds_batch": 22.393058201065287, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5420912504196167, "target_z": 0.9062500392599613, "ce": 6.0146708488464355, "genuineness": 4.507347106933594, "identity": 0.7378519773483276, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/038e9db747aa5cf08c60964dd9666e704d8b783a4c7f8456b5cb79fdae93559b.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "03923a8ad20c8c0b3087e4942430e8d897edd3aab909f2b7815f982a71440922", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "61072d457e4478afc189fb0fab40674127aad1496295e4dd46ada82d1d60e2dd", "generation_seconds_batch": 22.475721878930926, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.400529146194458, "target_z": 0.15234372546252423, "ce": 6.157639026641846, "genuineness": 4.482243061065674, "identity": 0.6139882802963257, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/03923a8ad20c8c0b3087e4942430e8d897edd3aab909f2b7815f982a71440922.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "039807bfe90603ed681b2ef6480b5fe204b59b270aff02e97de7e32c154fcfea", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "67d38ea3cf4a1a15a51c389cd1e66aec28ce32a4ba3d2a4dc9e646e79b29ec68", "generation_seconds_batch": 17.237184493918903, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1328877210617065, "target_z": 1.265625045803288, "ce": 5.510311126708984, "genuineness": 4.846253395080566, "identity": 0.6420234441757202, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/039807bfe90603ed681b2ef6480b5fe204b59b270aff02e97de7e32c154fcfea.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "03a2bcf4a489bbac6efbfc4b915e7d188c63e9d8b924a673444cc3c880db4ad1", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "61d19c4f92fd54d17061c585df57cd819575f40ea2ed2cb4ed641bcc330c3248", "generation_seconds_batch": 22.78277089993935, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6703099012374878, "target_z": 1.6718749803700195, "ce": 5.418970584869385, "genuineness": 4.883909225463867, "identity": 0.692151665687561, "hypothesis": "J'ai trouvé la petite boîte, l'été, 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/vn_STNC_reg_low/03a2bcf4a489bbac6efbfc4b915e7d188c63e9d8b924a673444cc3c880db4ad1.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "03a38668aeb443f8fba6e8cf3bc81ca12ad301e687ee421f7c4b56443b4e5548", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "119efa62575d40d5bd3c518a6c40ce8fdebcb02f242b3dbb031bfe3f73d839a1", "generation_seconds_batch": 22.716950829024427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.714788556098938, "target_z": 1.632812470555029, "ce": 5.137005805969238, "genuineness": 4.632867813110352, "identity": 0.572831392288208, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/03a38668aeb443f8fba6e8cf3bc81ca12ad301e687ee421f7c4b56443b4e5548.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "03c2639527b6be551eaea1007e97d3ca51684f90fa82f92a442f7b27eb635431", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d5449e81ce549967f68c4271df2b506696866431394a1d47d9832b9229bc3dd1", "generation_seconds_batch": 16.965796970995143, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.018012762069702, "target_z": 0.4882812679941489, "ce": 6.284721851348877, "genuineness": 4.68307638168335, "identity": 0.3312161862850189, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/03c2639527b6be551eaea1007e97d3ca51684f90fa82f92a442f7b27eb635431.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "0407cca8888b7212e2e769db95c987f7644ddd048dc5b195aed964d976398f24", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0d7706919d60309335d628af511e1485f252d33b9dc9c838eed1ba2ba812f7b2", "generation_seconds_batch": 21.393917390028946, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.631048560142517, "target_z": 0.8281250196299806, "ce": 6.157639026641846, "genuineness": 4.808597087860107, "identity": 0.7842046022415161, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/0407cca8888b7212e2e769db95c987f7644ddd048dc5b195aed964d976398f24.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "04600326215edf2552103151c6cc26d816b1e7731b292ced1e71bfa37a9b5acd", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "caecd9163914becbe77fdf0f007379a0988c0c49244102ed5a0bb96f09e052c2", "generation_seconds_batch": 17.0433306141058, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9601905345916748, "target_z": 0.5390625098149903, "ce": 5.919358730316162, "genuineness": 4.720732688903809, "identity": 0.6899203062057495, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/04600326215edf2552103151c6cc26d816b1e7731b292ced1e71bfa37a9b5acd.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": "047427617fc3cd7d04494aa4207faea5dfc596d7b2ea301f537cf535073eb94e", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1468b145cd4461ebbfc7698fa4e49200290662ce4d1eaad43e4945bf0306df44", "generation_seconds_batch": 16.01888156903442, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5865700244903564, "target_z": 0.8671874247517409, "ce": 5.6334228515625, "genuineness": 4.570107460021973, "identity": 0.6724289059638977, "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/vn_STNC_reg_low/047427617fc3cd7d04494aa4207faea5dfc596d7b2ea301f537cf535073eb94e.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "04d7b7cad7a655d72549cc1631c80d852081dd52c935422b8cea831e37edc9d1", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "efb6203ada3790829432ac6f80a8045cb3dbcefa20e9fa251f45fd6a714baa86", "generation_seconds_batch": 16.617920999997295, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.258197546005249, "target_z": 0.27734367311590924, "ce": 6.149696350097656, "genuineness": 4.545003414154053, "identity": 0.6428754925727844, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/04d7b7cad7a655d72549cc1631c80d852081dd52c935422b8cea831e37edc9d1.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "052a1ae9e08b1495dd5eb2c730f72cb10f1724ba2424bb10939058515522c268", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7a6920244b8d3c020e3e692bc20ff8ffe53b2d549bddf09a04cfcbe8adc9b509", "generation_seconds_batch": 22.377440257929265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.567323923110962, "target_z": 0.005859470696155539, "ce": 6.09409761428833, "genuineness": 4.557555198669434, "identity": 0.7157793045043945, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/052a1ae9e08b1495dd5eb2c730f72cb10f1724ba2424bb10939058515522c268.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": "05588f6b37d9e0fa1a56fc717c90b6d534f733f0648063dba33a3e5944a648e2", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8ecd14e4dada7a4dbf7d7883852c269b49488493d7c9571c8dba671609d55764", "generation_seconds_batch": 17.543925322010182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.275219440460205, "target_z": 1.1406249934566732, "ce": 5.621508598327637, "genuineness": 4.6203155517578125, "identity": 0.7908260822296143, "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/vn_STNC_reg_low/05588f6b37d9e0fa1a56fc717c90b6d534f733f0648063dba33a3e5944a648e2.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "d547d188d6308ff67ddbee697735986719cb74dc11a0dad13e4e8db27581311d", "generation_seconds_batch": 16.690406416077167, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5723278522491455, "target_z": 0.0014648676740388847, "ce": 6.109982967376709, "genuineness": 4.607763767242432, "identity": 0.5251220464706421, "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/vn_STNC_reg_low/059867605942d1f4c46c2c44d936e08b7fbe89dd4feef4c05500a7ed426c27a3.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "069f257b53ffaf8ccf488819dce6f7080b88df158e05e49e07d3dda1c5451f32", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "96e7707372a5c1e1ef6c947d6c76d082d4ed16a6630f7ba1a7fc7a31a9f60309", "generation_seconds_batch": 23.789345577941276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.631048560142517, "target_z": 0.8281250196299806, "ce": 6.062326908111572, "genuineness": 4.425758361816406, "identity": 0.5441396236419678, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/069f257b53ffaf8ccf488819dce6f7080b88df158e05e49e07d3dda1c5451f32.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "06e5efcc1c53826bd07d881b87c88d023a8969db2d66af1fb7c3936e4ccce528", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e52d74f877b9c2261f54e3cca3ebc1cde94cee2989b491e064761c98de5391b6", "generation_seconds_batch": 17.027500704978593, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.43534255027771, "target_z": 1.0, "ce": 6.1020402908325195, "genuineness": 4.65797233581543, "identity": 0.49833160638809204, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/06e5efcc1c53826bd07d881b87c88d023a8969db2d66af1fb7c3936e4ccce528.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "076b879cd64af2111a204ecf2023fb70a498c62e3f3294d4fbc4cdd4d187c0eb", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "88a258c9ade0271c0b0ad3da8ef7c88818b2498cd89bed417d5ded07cf65e560", "generation_seconds_batch": 17.343595108017325, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.426446795463562, "target_z": 1.0078125229016441, "ce": 5.25614595413208, "genuineness": 4.6203155517578125, "identity": 0.6427895426750183, "hypothesis": "La puerta sigue abierta. he dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/076b879cd64af2111a204ecf2023fb70a498c62e3f3294d4fbc4cdd4d187c0eb.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "078f364d318b72ee5c73881ccfdfba2b67f56d793aa8fc3a1e273af8f0a23f39", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d231755a5ef86579556c990549eceb26c302cf975de65b14f5a195224dcf1f24", "generation_seconds_batch": 17.65410869102925, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7377972602844238, "target_z": 0.7343750588899419, "ce": 6.252951145172119, "genuineness": 4.570107460021973, "identity": 0.4258473515510559, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/078f364d318b72ee5c73881ccfdfba2b67f56d793aa8fc3a1e273af8f0a23f39.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "087cb8194a68421979b3a418a89672edfa6b9600f948ac5af4e3311c8898e5aa", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "453483ccdc184879a2a882ea467e4402fa7e8cc045a0940037a58334586923e5", "generation_seconds_batch": 21.165586818940938, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6844229698181152, "target_z": 0.7812499869133462, "ce": 5.895530700683594, "genuineness": 4.758388519287109, "identity": 0.6844321489334106, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/087cb8194a68421979b3a418a89672edfa6b9600f948ac5af4e3311c8898e5aa.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": "08bc959326826731aa01459b2b9f9de5698423d4ddfb5b9c0039b12000f1b14e", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Someone moved the little blue chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a5179ff77b5f1f37d77f077149735b75731647490a653049652e7b5e5375232f", "generation_seconds_batch": 21.141588732018135, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.671079397201538, "target_z": 0.7929687189191974, "ce": 5.831989288330078, "genuineness": 4.381826400756836, "identity": 0.5023084878921509, "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/vn_STNC_reg_low/08bc959326826731aa01459b2b9f9de5698423d4ddfb5b9c0039b12000f1b14e.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "099952396eab4597f61560699683b540c72ed8881bcbe793a152bc4a5f21ca18", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "a2cd8c7e8f3a1e9728267dbd5c5b8261874a3f47030e241ce08516262a7d42fa", "generation_seconds_batch": 16.08077994803898, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9468469619750977, "target_z": 0.5507812418208414, "ce": 6.1020402908325195, "genuineness": 4.381826400756836, "identity": 0.576915979385376, "hypothesis": "The 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/vn_STNC_reg_low/099952396eab4597f61560699683b540c72ed8881bcbe793a152bc4a5f21ca18.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "09ae16d0743f4092b60f6b6f22adf6c46e85d7ea8647feb45e2292211d461d91", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e846c25b3d41d84319d1fbcd604f1fc97377c070541ebaf43da2e8f8e2f1acbc", "generation_seconds_batch": 24.461503827944398, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8749116659164429, "target_z": 1.4921874770983559, "ce": 5.498397350311279, "genuineness": 4.808597087860107, "identity": 0.8234262466430664, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/09ae16d0743f4092b60f6b6f22adf6c46e85d7ea8647feb45e2292211d461d91.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": "09b0a55f90d6c19988aaabe56dfe34fa1a0df9030b37cb3710e68d1927bb0186", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Someone moved the little blue chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "67df13bbe67810f63776b9063e77226d50a46a056b84b96e8a7c01821b69e60b", "generation_seconds_batch": 17.476938255946152, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8356503248214722, "target_z": 0.6484375163583171, "ce": 5.80021858215332, "genuineness": 4.457139015197754, "identity": 0.4780362844467163, "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/vn_STNC_reg_low/09b0a55f90d6c19988aaabe56dfe34fa1a0df9030b37cb3710e68d1927bb0186.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "09d9ffa69425903a22dd329a65c00ce245a09241c645cda74a48d8d15f01b778", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a951a9fefdc1bff5527111432531b00729d10267f927d970bc9b905d9de522fc", "generation_seconds_batch": 23.21326634590514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0691633224487305, "target_z": 0.4433592874830031, "ce": 5.919358730316162, "genuineness": 4.909013748168945, "identity": 0.6384388208389282, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/09d9ffa69425903a22dd329a65c00ce245a09241c645cda74a48d8d15f01b778.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "0a2adfe9f88d26b98d7da1e6ec32e93eec27cf6c91fd449e38651560b008c282", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c4d5fbd39849c54d1a0ea70930e1b59615f8f1072377e20e0b2f8fb59164fdfb", "generation_seconds_batch": 23.359947940101847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1069700717926025, "target_z": 0.4101562483641683, "ce": 5.486483573913574, "genuineness": 4.268857479095459, "identity": 0.5718734264373779, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/0a2adfe9f88d26b98d7da1e6ec32e93eec27cf6c91fd449e38651560b008c282.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "0a85e3862a398591dbb31d9212cbec4408377ced2e04bf2a46f8ed14f3806094", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "64c1b157bb61f459872bb963ed75293b147a41feb5daa323b9255e004b090934", "generation_seconds_batch": 22.75139091105666, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1062005758285522, "target_z": 1.2890625098149904, "ce": 5.442798614501953, "genuineness": 4.7458367347717285, "identity": 0.7721055746078491, "hypothesis": "Die Tür 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/vn_STNC_reg_low/0a85e3862a398591dbb31d9212cbec4408377ced2e04bf2a46f8ed14f3806094.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "0b0737878fe3de19224e11ec2258c0af189b03f9189e47d4f195a7e8c732a41f", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "513fb2e2186aec75d30dfd83f459553aa8c75505c360675a9c6f086cedd175e7", "generation_seconds_batch": 16.871221506036818, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8542113304138184, "target_z": -0.24609379089579297, "ce": 6.173524379730225, "genuineness": 4.118232727050781, "identity": 0.2846299707889557, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/0b0737878fe3de19224e11ec2258c0af189b03f9189e47d4f195a7e8c732a41f.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "0bbcda35691729c8c586f220a714d8dc245cb1fad7bf9f294717d4b3d8c14b8b", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "beff19d83b04c4614234b84514f5f696b51212d2eed55af08a593e3277e23350", "generation_seconds_batch": 22.535716219921596, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2396365404129028, "target_z": 1.1718749803700195, "ce": 5.863759994506836, "genuineness": 4.7081804275512695, "identity": 0.6486238241195679, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/0bbcda35691729c8c586f220a714d8dc245cb1fad7bf9f294717d4b3d8c14b8b.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "0bf34a3785bbac07ffdb1c61c11a0c8f77f0463bfea48b1b836e871787c0e64f", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "17e044788add1befbba9b9adbb054d6c53ed187e5e4e6afd980a53b5c0ad3b3b", "generation_seconds_batch": 22.68292859394569, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8445460796356201, "target_z": 0.6406249934566731, "ce": 5.967014789581299, "genuineness": 4.796044826507568, "identity": 0.7500381469726562, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/0bf34a3785bbac07ffdb1c61c11a0c8f77f0463bfea48b1b836e871787c0e64f.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "0c03dea5a812c818d00a0731f25c51bc734d149157c0c9c481d8d162ed18d373", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4a80acb03f0efcec4867688cec3af38830f57d9765c15abf2efebd298323fa72", "generation_seconds_batch": 16.62918314000126, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.675184726715088, "target_z": -0.08886717318647246, "ce": 5.816103935241699, "genuineness": 4.632867813110352, "identity": 0.7970398664474487, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/0c03dea5a812c818d00a0731f25c51bc734d149157c0c9c481d8d162ed18d373.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "0c31aef4cfc22e2de99a20e5000cbc5ad4885138cc57e359309272b11a727235", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a1c60b65e47353e971f9ef13daf552106e8e304adfac3e35185bc24fe5eb87c0", "generation_seconds_batch": 17.91963773907628, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4531340599060059, "target_z": 0.9843749541967118, "ce": 5.990842819213867, "genuineness": 4.7332844734191895, "identity": 0.3672642707824707, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. wanneer je klaar bent.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/0c31aef4cfc22e2de99a20e5000cbc5ad4885138cc57e359309272b11a727235.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "0c342a66bcead5e845a606f1466cae4360618c28337209c6090dcf61aa83420f", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2ec809de9a503d35b19556c32256c2f8711b94d42c3cbcd744a0197635978dfe", "generation_seconds_batch": 21.561668646987528, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.601794958114624, "target_z": -0.024414042461061446, "ce": 6.062326908111572, "genuineness": 4.519899368286133, "identity": 0.7313183546066284, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/0c342a66bcead5e845a606f1466cae4360618c28337209c6090dcf61aa83420f.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "0c701c5150127f666512d86992ad9a12092eab6892e180b1f3429d4790b4f990", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ffd0cfb30280b618e2e8b7bdb5f782f89fe0edf42af80ad388407561d4779a34", "generation_seconds_batch": 17.34895976597909, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.147000789642334, "target_z": 0.375000052346615, "ce": 6.1020402908325195, "genuineness": 4.68307638168335, "identity": 0.5140575170516968, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/0c701c5150127f666512d86992ad9a12092eab6892e180b1f3429d4790b4f990.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "0c709190ef0d0e9956083b7243edcdc19f825f424df6a387d48c0f79169b362d", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5e12604366da5d00c11cc2043183949e167020fb3cfc83292897b17a569a1543", "generation_seconds_batch": 16.177126638009213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3994171619415283, "target_z": 0.15332030391188348, "ce": 6.300607204437256, "genuineness": 4.231201171875, "identity": 0.577155590057373, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/0c709190ef0d0e9956083b7243edcdc19f825f424df6a387d48c0f79169b362d.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": "0c9c035e5d4c50129b811d99e1b994df77ad77236a5174f5b8c8c8954ba0b405", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "bcdd43503f7d9d3ad63460edb758649538e6ee3d121eb2dd6b1c21a52419beeb", "generation_seconds_batch": 16.90205033204984, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1506792306900024, "target_z": 1.25, "ce": 5.470598220825195, "genuineness": 4.557555198669434, "identity": 0.5665509700775146, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna, naast het raam, neem ze mee als je weggaat.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/0c9c035e5d4c50129b811d99e1b994df77ad77236a5174f5b8c8c8954ba0b405.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": "0cb02b1966b63adcdb9e8d499fc7c7a2fd0a0c489dd7c1cd1981ecb418b2cd52", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "27d0571416d11e0729d61e44709c27373f39980d20c3d88f71999e4c045813f1", "generation_seconds_batch": 16.98819609102793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2307407855987549, "target_z": 1.1796875032716634, "ce": 5.748591423034668, "genuineness": 4.595211505889893, "identity": 0.7056360244750977, "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/vn_STNC_reg_low/0cb02b1966b63adcdb9e8d499fc7c7a2fd0a0c489dd7c1cd1981ecb418b2cd52.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "0cb2f01a38401aed44b1ed5ff24bd781b1a7acbee5f10d7599156811c170054a", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bb037426d5e02aa7bacf2abacd4f9d37a7d271a5e3d18ed239f16887657a27c8", "generation_seconds_batch": 23.196196221979335, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1150963306427002, "target_z": 1.2812499869133462, "ce": 5.585766792297363, "genuineness": 4.6203155517578125, "identity": 0.768258273601532, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/0cb2f01a38401aed44b1ed5ff24bd781b1a7acbee5f10d7599156811c170054a.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "adeb7519263c604fb05dc5eb59de0658a81f30b5b351ac894d30e03c5a158a25", "generation_seconds_batch": 17.001083524082787, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.178905487060547, "target_z": -0.5312500916065762, "ce": 6.086154937744141, "genuineness": 4.325342178344727, "identity": 0.8217686414718628, "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/vn_STNC_reg_low/0cb70e86df8754d3bba44fe7f798aac7d0ffe2ca68fc75b92ff1a1ed29986ba7.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "0d6b17c9c0370312e14a703fc9a4a8da4e17eadab86d5e7731fbbe1f7caaacfb", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a64c36e65f2d7dd393b0cad7913800369ac700ad569ca97aab27e307199e1a54", "generation_seconds_batch": 16.61231451795902, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1425528526306152, "target_z": 0.37890636614405204, "ce": 5.919358730316162, "genuineness": 4.306513786315918, "identity": 0.7015982270240784, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/0d6b17c9c0370312e14a703fc9a4a8da4e17eadab86d5e7731fbbe1f7caaacfb.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "0d9c27a9d3e6306f90d3e7eb48ede3f35ed3933241611c509ff4c7290411afab", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "03e7285d05ce3b59750ee669e80725af5e2c8f0c717d223f82bcedfe370d65ce", "generation_seconds_batch": 22.418045150930993, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7733802795410156, "target_z": 0.7031249672833656, "ce": 6.1020402908325195, "genuineness": 4.632867813110352, "identity": 0.7020857334136963, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/0d9c27a9d3e6306f90d3e7eb48ede3f35ed3933241611c509ff4c7290411afab.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "0dc0925557b4f876131e12617aae4bcf5513f5af6944b4bf22da91633fbfe6da", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "21cf8187551758b16830471491e559383fabf3137e3593598e6db18c59710d13", "generation_seconds_batch": 17.952876089024357, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7200058698654175, "target_z": 0.75, "ce": 6.078212261199951, "genuineness": 4.7834930419921875, "identity": 0.7022513747215271, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/0dc0925557b4f876131e12617aae4bcf5513f5af6944b4bf22da91633fbfe6da.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "0e8dfa53da68c3bfd21eec95394fa349155906b10ff7f1207bcb8d3183a976ef", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "07e0887173b54b955ef655000102dfce0bfdb3aded9853c3ba6f0c332f49ccad", "generation_seconds_batch": 22.964052592986263, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8126415014266968, "target_z": 1.5468750327166343, "ce": 5.855817794799805, "genuineness": 4.582659721374512, "identity": 0.44875141978263855, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/0e8dfa53da68c3bfd21eec95394fa349155906b10ff7f1207bcb8d3183a976ef.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "0f1a2ed1d68b576faf084f3f0cefef64e4c23b3c83778e8eb5eca39c3bbc6ea0", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "58c409b5a802aec0b0482806be5d699e18048555a0d1aa89d29a0eb49ef3a079", "generation_seconds_batch": 23.359947940101847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7111101150512695, "target_z": 0.757812522901644, "ce": 5.446769714355469, "genuineness": 4.394378662109375, "identity": 0.48093926906585693, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/0f1a2ed1d68b576faf084f3f0cefef64e4c23b3c83778e8eb5eca39c3bbc6ea0.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "0f3c2a4951ce8dc3efd52f94d5b32445d6656030952e773f68bb13df7f34a046", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "729501dbb1693aac8d7ed572fdb93a6301b0a24b1861ce2865df4948ce2b8d8c", "generation_seconds_batch": 24.293172053061426, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1506792306900024, "target_z": 1.25, "ce": 5.510311126708984, "genuineness": 4.7081804275512695, "identity": 0.590511679649353, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/0f3c2a4951ce8dc3efd52f94d5b32445d6656030952e773f68bb13df7f34a046.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "0f769a084fe9b822578e99b9f10caa2db91b917b6eacb8618cd720b70a85a268", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1aeec08c4f09e413d06e8833185c7d81cc675f602d6861cad4ece51df0b5dbb5", "generation_seconds_batch": 17.11507529800292, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0802829265594482, "target_z": 0.43359371237587047, "ce": 6.41180419921875, "genuineness": 4.7081804275512695, "identity": 0.4976566433906555, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/0f769a084fe9b822578e99b9f10caa2db91b917b6eacb8618cd720b70a85a268.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": "0f93f6ca104a18313011cf7e4255abc2cd38fcf059f7fa7bd910d390e7e5e2a7", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e907b3a880f1389419e9123e3c48a83777fa465a405399c3a9716266d0a371d2", "generation_seconds_batch": 16.28377297706902, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1239920854568481, "target_z": 1.273437464011702, "ce": 5.720792293548584, "genuineness": 4.796044826507568, "identity": 0.8181661367416382, "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/vn_STNC_reg_low/0f93f6ca104a18313011cf7e4255abc2cd38fcf059f7fa7bd910d390e7e5e2a7.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "0f9cd9dda293e0817514a4fada6d7e489d8ce5693b48e54ee2b00f5b6e39698f", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ab103b19e4309bb135aa2b22c6daf42a8858e4aa244f5670371c84e63f3a0782", "generation_seconds_batch": 23.500529718003236, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.648840069770813, "target_z": 0.8124999738266925, "ce": 6.117925643920898, "genuineness": 4.758388519287109, "identity": 0.6435844898223877, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/0f9cd9dda293e0817514a4fada6d7e489d8ce5693b48e54ee2b00f5b6e39698f.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "0fc3c8b5a313f108d2a32d18eefdfe88b7673697f3630d4676cca90f02affa6b", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ba98d113abd0b7d931a17896125999d2ac85e026c9782aa973eeac43560213c6", "generation_seconds_batch": 21.9917241219664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1684707403182983, "target_z": 1.234374954196712, "ce": 5.764476776123047, "genuineness": 4.833701133728027, "identity": 0.767308235168457, "hypothesis": "La porte est La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.16666666666666666, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/0fc3c8b5a313f108d2a32d18eefdfe88b7673697f3630d4676cca90f02affa6b.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "0fe5f28a37b0de044e805c6d330e8260014675105e59983628247534c6644d90", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "4c4e3097095a4729ca207b61860ba3f2742939eb635e2d204f07a1813be60e03", "generation_seconds_batch": 17.118609820026904, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0350347757339478, "target_z": 1.3515624836416829, "ce": 6.109982967376709, "genuineness": 4.645420074462891, "identity": 0.38888099789619446, "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/vn_STNC_reg_low/0fe5f28a37b0de044e805c6d330e8260014675105e59983628247534c6644d90.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": "0ff6404023abb11ca11d13af9313c8636a080c841d9ae0fcdd1b32af4889e4ec", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "60bd660466c2aee0c9f2f0648c06526c20e173924e27ca8c1a74f1b02a4bc3c6", "generation_seconds_batch": 20.705067010014318, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2396365404129028, "target_z": 1.1718749803700195, "ce": 5.514282703399658, "genuineness": 4.68307638168335, "identity": 0.670928955078125, "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/vn_STNC_reg_low/0ff6404023abb11ca11d13af9313c8636a080c841d9ae0fcdd1b32af4889e4ec.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "100452c4e22aa7ec6897959eade809c915472307d2577720362bc00220739f05", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ca8ec5f109226765c6fd74a2c7e68c8dd7e6afae249694e7d687a9f05cb5ebd1", "generation_seconds_batch": 21.743415127974004, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.444238305091858, "target_z": 0.992187477098356, "ce": 5.692992687225342, "genuineness": 4.494794845581055, "identity": 0.25905150175094604, "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/vn_STNC_reg_low/100452c4e22aa7ec6897959eade809c915472307d2577720362bc00220739f05.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "100b553715d929d5c387bff406894430c6c934b445f3fc4caf2c9b0d8290d010", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2cb9ecfecfa4da2af788bb7a22d811abbdf87e177750a0ce08a6a37914bca1f4", "generation_seconds_batch": 17.23874403501395, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.466135025024414, "target_z": 0.094726643882628, "ce": 6.221180438995361, "genuineness": 4.494794845581055, "identity": 0.714275598526001, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/100b553715d929d5c387bff406894430c6c934b445f3fc4caf2c9b0d8290d010.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "10c5c8ab33d45c60fcc5d59320773785d14f71d306b6bfd34c39db6cdfe6994c", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "68639d06407494a582776c7aed752a3bf70fe85cab03dbb4bb54291eb7406490", "generation_seconds_batch": 23.733957962016575, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.417551040649414, "target_z": 1.015625045803288, "ce": 5.982900142669678, "genuineness": 4.532451152801514, "identity": 0.44459009170532227, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/10c5c8ab33d45c60fcc5d59320773785d14f71d306b6bfd34c39db6cdfe6994c.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "10ec58b683b4d26a3daf2eb8fd9196975f5c66b1f7f06483803c578edc8a802c", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f57ab5e43df221e2b0272e33e71cd68e294c9783a07402263e6412a8d74bdc6f", "generation_seconds_batch": 16.301957082003355, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1951578855514526, "target_z": 1.2109374901850096, "ce": 5.085378646850586, "genuineness": 4.507347106933594, "identity": 0.7909239530563354, "hypothesis": "J'ai trouvé la 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/vn_STNC_reg_low/10ec58b683b4d26a3daf2eb8fd9196975f5c66b1f7f06483803c578edc8a802c.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "114bcab098da87dd2d684ff8bb73c27638a5979c1e85af19490918ab88688507", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "44ca56fb83732a5b47ad8b278e4b4d4ee5190bc007e673dc94d83f090751017f", "generation_seconds_batch": 16.563936572987586, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.240406036376953, "target_z": 0.29296871891919735, "ce": 5.824047088623047, "genuineness": 4.3881025314331055, "identity": 0.5405558943748474, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/114bcab098da87dd2d684ff8bb73c27638a5979c1e85af19490918ab88688507.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "127a9c1d10c01ca6654e4d27e05d2ba226aab4712ff91217e24513448624d9a9", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "beb94351d19e5f22bac77c6b065b6bb7d14b8e558d952f66e7f41beac3bdd313", "generation_seconds_batch": 22.45146721205674, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.995773434638977, "target_z": 0.507812522901644, "ce": 6.054384231567383, "genuineness": 4.582659721374512, "identity": 0.6126692891120911, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/127a9c1d10c01ca6654e4d27e05d2ba226aab4712ff91217e24513448624d9a9.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "0f99eb66ce2e78a6916f18c571443446b02c28c72c0f513b64f059eb23ea8765", "generation_seconds_batch": 17.819578128983267, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6551692485809326, "target_z": -0.07128897048446582, "ce": 6.284721851348877, "genuineness": 4.444586753845215, "identity": 0.7068289518356323, "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/vn_STNC_reg_low/12a7b850c68b1e935d69d2b179f3c0df54c4775061935760902fd9f6cbfc7541.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "12b99e327ea16e94bb035457684141c9f64d5eaf1ccc004a06c13f3fd8eada5c", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c00d8f0efc8ef7ede421da352d6f7959857f58c6f96beafab0ca4cc3dfd5fcac", "generation_seconds_batch": 22.017901215003803, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.791171669960022, "target_z": 0.6875000261733075, "ce": 5.967014789581299, "genuineness": 4.570107460021973, "identity": 0.6214962005615234, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/12b99e327ea16e94bb035457684141c9f64d5eaf1ccc004a06c13f3fd8eada5c.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "12c108cd17d250f9abed2864fcc90c26f6c1dc6406a336891cd8467f8f01f06d", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "83c4d29f62bf575b8e0765f3ec5ca194bd4e835984ce2bac87ee3b17b050c47d", "generation_seconds_batch": 22.449539219029248, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7733802795410156, "target_z": 0.7031249672833656, "ce": 6.157639026641846, "genuineness": 4.821148872375488, "identity": 0.7278902530670166, "hypothesis": "J'ai trouvé la bah c'est trouvé, la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.21052631578947367, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/12c108cd17d250f9abed2864fcc90c26f6c1dc6406a336891cd8467f8f01f06d.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "12d5133d7aa19736d87f28b4db46d7904b293b3402d39800c19f70804246c194", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2e26e90319cc2f8e7c16a9c0a8334b8eb7e0171bcfdbc8f586fc155ecd070f8d", "generation_seconds_batch": 16.26477698306553, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5420912504196167, "target_z": 0.9062500392599613, "ce": 5.569881439208984, "genuineness": 4.68307638168335, "identity": 0.47623211145401, "hypothesis": "He encontrado la cajita. estaba junto a la ventana vija. ¿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/vn_STNC_reg_low/12d5133d7aa19736d87f28b4db46d7904b293b3402d39800c19f70804246c194.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "12d5cf3720f075a8e3ae7676e800d963d915f1fd18ff342ba144beba8e803502", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "264d336b73dd933972110376e73916b80cbc2c6891f88821fda86af02a3aef25", "generation_seconds_batch": 23.679422381101176, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.042476177215576, "target_z": 0.46679675149470523, "ce": 5.959072113037109, "genuineness": 4.381826400756836, "identity": 0.1830979287624359, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/12d5cf3720f075a8e3ae7676e800d963d915f1fd18ff342ba144beba8e803502.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "12e5695cc542ec82899b074ae43204ad6b2cadfd2652a43b3efb4a27d9335003", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "76fcb0290eaef0f583568c3e2a33fcb566aab0972b408b62f43390f5b46728b1", "generation_seconds_batch": 16.598701850976795, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0261390209197998, "target_z": 1.3593750065433268, "ce": 5.577824115753174, "genuineness": 4.607763767242432, "identity": 0.6827417016029358, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/12e5695cc542ec82899b074ae43204ad6b2cadfd2652a43b3efb4a27d9335003.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "1304cabc2f8479a4255f9905b9ede518ea85b9f3690da2033c9edc385d262d0d", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "082b9e71ba6cdf32a093a4e124de87828706b144df59f8cc73a433897e51583c", "generation_seconds_batch": 23.705427067005076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8089631795883179, "target_z": 0.6718749803700194, "ce": 5.824047088623047, "genuineness": 4.570107460021973, "identity": 0.7116976380348206, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/1304cabc2f8479a4255f9905b9ede518ea85b9f3690da2033c9edc385d262d0d.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "131f242618205b937c83118f7ee0b3f870806995ceba6620b00fafce1c27a493", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4a9fa541d589e80bbc3b5fb85f6044a314c21a5403cdcb8429c76b1d8b721d4f", "generation_seconds_batch": 18.615822600084357, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8475394248962402, "target_z": -0.24023432019963742, "ce": 5.95112943649292, "genuineness": 4.457139015197754, "identity": 0.6392887830734253, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/131f242618205b937c83118f7ee0b3f870806995ceba6620b00fafce1c27a493.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "13a79edc5ab4ba1d7bab598b1d666d93ec14a30b9adb4186c212fca03ee96103", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cb45220d34739525e235e78c716bbbf985bb17f0afb22995b697431f0fb0cf74", "generation_seconds_batch": 22.682070187991485, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1803598403930664, "target_z": 0.3457031176387573, "ce": 6.1894097328186035, "genuineness": 4.6203155517578125, "identity": 0.5069008469581604, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/13a79edc5ab4ba1d7bab598b1d666d93ec14a30b9adb4186c212fca03ee96103.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": "14591d68665fb180665e39a9038ea17faf4c55a607e0e054d426fea21de8b0c5", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "50f5faa5c9bf421ec09ec9986f5761091ce997c13c3e5af790feaff4a092ed69", "generation_seconds_batch": 16.77914552600123, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3641767501831055, "target_z": 1.0624999738266925, "ce": 5.637393951416016, "genuineness": 4.6705241203308105, "identity": 0.680047869682312, "hypothesis": "Las llaves están en la mesa de la cocina, miren el cajón y luego junto a la ventana, llévatelas cuando te vayas.", "wer": 0.08695652173913043, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/14591d68665fb180665e39a9038ea17faf4c55a607e0e054d426fea21de8b0c5.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "14961a5885386f71abe996501e45a22e90041902130bb8c55a6971181bd20c80", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1eb9454feb9d62101be02649098dd66d3ce17f47beb8d83028efd233982933af", "generation_seconds_batch": 23.34825802803971, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1773663759231567, "target_z": 1.226562535988298, "ce": 5.34748649597168, "genuineness": 4.595211505889893, "identity": 0.4615226686000824, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/14961a5885386f71abe996501e45a22e90041902130bb8c55a6971181bd20c80.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "14fb3ff7868a73cb5a35e8bd748a466e2197a9d3465e606b3920de355ee19325", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fdf08e2475f1d2518b30f1b256174449293c2b169b06b617d05241723b22f51d", "generation_seconds_batch": 22.790029660100117, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1595749855041504, "target_z": 1.2421874770983559, "ce": 5.502368927001953, "genuineness": 4.6705241203308105, "identity": 0.6440587043762207, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/14fb3ff7868a73cb5a35e8bd748a466e2197a9d3465e606b3920de355ee19325.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "152c0a04d648f977ae2419cf36f876972376667c371897f8371c6fe15a093dab", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e1effbac0578348d860b85548157a8780c4b20485b892b6154a85864ecbc68ea", "generation_seconds_batch": 22.336110053001903, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2396365404129028, "target_z": 1.1718749803700195, "ce": 5.538110733032227, "genuineness": 4.519899368286133, "identity": 0.6340979337692261, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/152c0a04d648f977ae2419cf36f876972376667c371897f8371c6fe15a093dab.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aca85e0be0a0e039c0b25ee5f56829d240644996036671226db9d84d8256f621", "generation_seconds_batch": 16.59049503400456, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.761918067932129, "target_z": -0.16503903591773458, "ce": 6.356205940246582, "genuineness": 4.419482707977295, "identity": 0.44489407539367676, "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/vn_STNC_reg_low/15585f3962d31fd159bb39a755593ab24aa0df87e4725d7772f40811258760f5.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "15650425d3022dd2b744da3558d91b44c7adcf25094d482a85e03772d14b98b2", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2e29d6c61d856317c79300d136cc020b409ab74667332f403a69f1e4bb707c9f", "generation_seconds_batch": 22.56491175701376, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6357100009918213, "target_z": -0.0541992663935988, "ce": 6.133810997009277, "genuineness": 4.720732688903809, "identity": 0.7371788024902344, "hypothesis": "La porte est encore ouverte. J'ai posé ton 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/vn_STNC_reg_low/15650425d3022dd2b744da3558d91b44c7adcf25094d482a85e03772d14b98b2.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "1574b88cc3f47086ab02a11d99d6bada3ce2652bbc79a697434e684d83feec63", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "473286ce51416d6787ddb581c1e112b7e70a0be71cbd1eaa107dba6cf16fbff0", "generation_seconds_batch": 22.393712619901635, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.942399024963379, "target_z": 0.5546875556182784, "ce": 6.221180438995361, "genuineness": 4.65797233581543, "identity": 0.329887330532074, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/1574b88cc3f47086ab02a11d99d6bada3ce2652bbc79a697434e684d83feec63.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "1596eb74a91690768291f01d39017b132c9faf399220c0baa093a9892bbcf5eb", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "acbd7f6147413a907b7e20295f167df8d4fee0a7644bb2855044324c88da27ef", "generation_seconds_batch": 17.96664779097773, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.382737636566162, "target_z": 0.16796877126581233, "ce": 6.086154937744141, "genuineness": 4.6956281661987305, "identity": 0.66612708568573, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/1596eb74a91690768291f01d39017b132c9faf399220c0baa093a9892bbcf5eb.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "160c6728ec5297a381a61e6f6c5e5bbd2e6534051b28bbcb1aedc82c234773a0", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d1e4d630cf2c45b5c5b494c9110a4101c618236bcaadbd01d90266f56cb70954", "generation_seconds_batch": 22.315808710991405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5687785148620605, "target_z": 0.8828124705550291, "ce": 5.7684478759765625, "genuineness": 4.720732688903809, "identity": 0.7287673950195312, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/160c6728ec5297a381a61e6f6c5e5bbd2e6534051b28bbcb1aedc82c234773a0.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "16a29f2cbdbb957455b9b50350ac669784f01e6157f5938b7f6ecce32b238cfc", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "47c42c1e5fe07376de446fdd0f743053d79bfbd40e3457065037159826e5524c", "generation_seconds_batch": 16.675345901981927, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.838643789291382, "target_z": -0.23242190199122337, "ce": 6.41180419921875, "genuineness": 4.419482707977295, "identity": 0.6407196521759033, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/16a29f2cbdbb957455b9b50350ac669784f01e6157f5938b7f6ecce32b238cfc.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "16f0216d3f668eb94dcd15e06f879da20a241211ed8591e78c3b113750cc1d3d", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "787b78ff5a960bfda1e53bdaa5087d9bc7683c0096bdd92e6118487843bde8f9", "generation_seconds_batch": 17.237184493918903, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3285938501358032, "target_z": 1.0937499607400387, "ce": 5.498397350311279, "genuineness": 4.570107460021973, "identity": 0.6943152546882629, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/16f0216d3f668eb94dcd15e06f879da20a241211ed8591e78c3b113750cc1d3d.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "16f9dcdc1192e1184953f5f4ce5c092eb19e8ddfb75b8480b6399096e9672ddb", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "415c7a99c220d8115b96f5dace829bd103fe8e686c08dc988fee931fb44b83ee", "generation_seconds_batch": 17.040524665964767, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9409446716308594, "target_z": -0.32226565362705506, "ce": 6.0146708488464355, "genuineness": 4.256305694580078, "identity": 0.4806883931159973, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/16f9dcdc1192e1184953f5f4ce5c092eb19e8ddfb75b8480b6399096e9672ddb.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "17cdeb0b10b3b138684128658234cc7612b05c5465596cc2810bc904c2d37e33", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ded30435a740949143bc8f6a38106bc24e09aab7aa0a14db5659e81b1d18f66f", "generation_seconds_batch": 23.12531906401273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7681628465652466, "target_z": 1.5859375425316247, "ce": 5.681078910827637, "genuineness": 4.545003414154053, "identity": 0.2611830532550812, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/17cdeb0b10b3b138684128658234cc7612b05c5465596cc2810bc904c2d37e33.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "181cbadbfe20301e2fc7cee1e129917565178ff7ab948f46ff46796e8191db0a", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "772c85dcf734829b125177f3b42345e3e0b93303773f31b574e0bbffcec78f8d", "generation_seconds_batch": 23.148726623039693, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5821220874786377, "target_z": 0.8710937385491779, "ce": 6.062326908111572, "genuineness": 4.65797233581543, "identity": 0.6701570749282837, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/181cbadbfe20301e2fc7cee1e129917565178ff7ab948f46ff46796e8191db0a.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "183361628df30d2d83cb2d0a6d8cc90a4684016e921bd4489f66cf593293c7ff", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a7308f5f396ef1e4466e861fc7eb9bab8ea7e522a87900bfec4ea382184b0266", "generation_seconds_batch": 23.42693795298692, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.07138729095459, "target_z": 0.4414061305842845, "ce": 5.95112943649292, "genuineness": 4.457139015197754, "identity": 0.5942560434341431, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Wanneer je klaar bent.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/183361628df30d2d83cb2d0a6d8cc90a4684016e921bd4489f66cf593293c7ff.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "188a1e437b6e4b1fa3d95bd7fb0d657e526d7407a5b6091f4e62916e8c1bd466", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "423a892a5c7f3ce76240afe4437407de0f98ff83fd6d8eccce121e3ca3ad3649", "generation_seconds_batch": 22.711376299965195, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.417551040649414, "target_z": 1.015625045803288, "ce": 6.181467056274414, "genuineness": 4.65797233581543, "identity": 0.7074193358421326, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/188a1e437b6e4b1fa3d95bd7fb0d657e526d7407a5b6091f4e62916e8c1bd466.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "18913330c8c63e291a7481a5fb67d09f2935a6a5d3a6391d9dbf71327108b52d", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c86b9343f6e3cc797f6521776c24374fe9ca305afd6a7527f2dc88800e873809", "generation_seconds_batch": 21.8722267660778, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0261390209197998, "target_z": 1.3593750065433268, "ce": 5.903473377227783, "genuineness": 4.482243061065674, "identity": 0.7459022998809814, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/18913330c8c63e291a7481a5fb67d09f2935a6a5d3a6391d9dbf71327108b52d.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "195b0c72b6233e4b05ccc9789e08012e65cdd2050f48a929d1a9dd50a40f7bd8", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dd796210ede3d378fcb277ff98baa0420449c15a39cd489fb1731e9377c96c7e", "generation_seconds_batch": 17.098108747974038, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8230762481689453, "target_z": -0.21875001308665376, "ce": 6.054384231567383, "genuineness": 4.607763767242432, "identity": 0.6046499609947205, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/195b0c72b6233e4b05ccc9789e08012e65cdd2050f48a929d1a9dd50a40f7bd8.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "1a032d5be50bbccec66f5bbc00fa1c000cb798ccce546346fa8b1c692dbbc416", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f01854433fb4da5f38cd18343c549d80aa011c552e9c1d55755238e52827584c", "generation_seconds_batch": 23.34184319106862, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.444238305091858, "target_z": 0.992187477098356, "ce": 5.704906940460205, "genuineness": 4.7332844734191895, "identity": 0.744055449962616, "hypothesis": "Die Tür 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/vn_STNC_reg_low/1a032d5be50bbccec66f5bbc00fa1c000cb798ccce546346fa8b1c692dbbc416.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "1a12935bd02124a1fc6c014ef4ea7a7da7dd7d579b0d1275cbbde20d33f2b3a1", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5fb7322145f9042ab05f0a94344840e9260acea9bf412abd97893cfde02e9fb0", "generation_seconds_batch": 22.87628270604182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0002212524414062, "target_z": 0.503906313797437, "ce": 5.601652145385742, "genuineness": 4.457139015197754, "identity": 0.499103307723999, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/1a12935bd02124a1fc6c014ef4ea7a7da7dd7d579b0d1275cbbde20d33f2b3a1.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "1aff43920d2822dcdba9f8b1d483c50d5d6c187b2fd8deb45cba3e3ffd502984", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "36513b3f03ca12b1262c2aa8d8552d151a1c055dd12da0c2132ecf3ac92776e4", "generation_seconds_batch": 23.500529718003236, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6562812328338623, "target_z": -0.07226554893382509, "ce": 6.165581703186035, "genuineness": 4.507347106933594, "identity": 0.7527003288269043, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/1aff43920d2822dcdba9f8b1d483c50d5d6c187b2fd8deb45cba3e3ffd502984.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "1b14be99ef3bb5a8dcd180452376c4bd7e54d55dc2365628fd1a12ec5b455fd3", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "26141403797f5ede58b4508b78b58580cb49a2c3954b385da1b115d4bfb4d41b", "generation_seconds_batch": 21.43839480006136, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9015988111495972, "target_z": 1.4687500130866538, "ce": 5.816103935241699, "genuineness": 4.545003414154053, "identity": 0.520398736000061, "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/vn_STNC_reg_low/1b14be99ef3bb5a8dcd180452376c4bd7e54d55dc2365628fd1a12ec5b455fd3.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "1b48c18a2eb96a9b80722c8f83ad45b7b296b653d67e2cda495c6b993e2799b1", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a6f136fcd52ae799e77781e0a9f10fd2a5d4bd3c58d9273bc1c6013aa24d939e", "generation_seconds_batch": 22.393058201065287, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0261390209197998, "target_z": 1.3593750065433268, "ce": 5.621508598327637, "genuineness": 4.607763767242432, "identity": 0.1583375632762909, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/1b48c18a2eb96a9b80722c8f83ad45b7b296b653d67e2cda495c6b993e2799b1.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "1b6a53b1a2280dce3c0116914168d799c395599f7a845133ac2a312cceb2b25b", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "618657411a49af9b9e8215d0d0220d12be260245a77491caffcc573d36a3b596", "generation_seconds_batch": 17.201869227108546, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9023683071136475, "target_z": 0.5898437516358317, "ce": 5.990842819213867, "genuineness": 4.406930446624756, "identity": 0.5588996410369873, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/1b6a53b1a2280dce3c0116914168d799c395599f7a845133ac2a312cceb2b25b.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "1b799ab2c151901e547691be3feb8a138b7b8590d95a9edf43f16a685115b87a", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5350af46aed7e8d387f95ce2756082d88f64c6060e070c06d6053e55264df946", "generation_seconds_batch": 17.175231553963386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6221528053283691, "target_z": 0.8359375425316247, "ce": 6.062326908111572, "genuineness": 4.909013748168945, "identity": 0.5853660702705383, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/1b799ab2c151901e547691be3feb8a138b7b8590d95a9edf43f16a685115b87a.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "1b9a16fb64611c6c55732f776c4032775ee31ae409ff8664ad92cde52ee56302", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5a3e247de2e9e3c30989f5370481eba431cda69492809d330450765731dff801", "generation_seconds_batch": 17.65410869102925, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5065083503723145, "target_z": 0.9375000261733075, "ce": 6.284721851348877, "genuineness": 4.7834930419921875, "identity": 0.3895722031593323, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/1b9a16fb64611c6c55732f776c4032775ee31ae409ff8664ad92cde52ee56302.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "1bacb535a330089b6cd031516df151618a4415c7cd591b27e3d4b14046b8e3a8", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d8a9e8497ef5f15dda4579eea4a6be40423a632a6c6bfc94079a3b196917897f", "generation_seconds_batch": 22.790029660100117, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.061721920967102, "target_z": 1.3281250196299805, "ce": 5.685050010681152, "genuineness": 4.758388519287109, "identity": 0.6013613343238831, "hypothesis": "La puerta sigue abierta. he dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/1bacb535a330089b6cd031516df151618a4415c7cd591b27e3d4b14046b8e3a8.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "1ce37074160d342891f4e128ab0aff4241a9c8f1b52b888305a7126e02a02bf6", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2b76164d76f8b5650abaef880003c77402248810483bd5f5bd9f237d8bc8fbfc", "generation_seconds_batch": 22.782622253056616, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3908638954162598, "target_z": 1.0390625098149904, "ce": 5.673136234283447, "genuineness": 4.720732688903809, "identity": 0.6783808469772339, "hypothesis": "He encontrado la cajita, estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/1ce37074160d342891f4e128ab0aff4241a9c8f1b52b888305a7126e02a02bf6.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "1ce58344982f14321ed18e922dc442711f9ceed20c18cb997a7fff6e37e40bfe", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "53940a99b3cf0550b0b87d955454e8dbef2e3589f0e69840feb1869b75c95a33", "generation_seconds_batch": 17.96664779097773, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.049147844314575, "target_z": 0.4609374901850097, "ce": 5.824047088623047, "genuineness": 4.494794845581055, "identity": 0.5921521186828613, "hypothesis": "J'ai trouvé la petite boîte, elle était près de de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/1ce58344982f14321ed18e922dc442711f9ceed20c18cb997a7fff6e37e40bfe.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "1cf5e84177f5033fa294dfe75839cf160d9e9a1dc98b8f41b117d7db221fa945", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "71dee4c287e02ca34c759d3d852381d5fef9993be55b192307e0d68f545d7e93", "generation_seconds_batch": 23.60449260403402, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5598827600479126, "target_z": 0.8906249934566731, "ce": 6.070269584655762, "genuineness": 4.507347106933594, "identity": 0.2982199788093567, "hypothesis": "I found the little box. I it was beside the old window. We can look inside together", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/1cf5e84177f5033fa294dfe75839cf160d9e9a1dc98b8f41b117d7db221fa945.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "71fbe081a0e29c82ce3d31aee62b5e4f1d3601a60ff24d5345e105ea54fedb64", "generation_seconds_batch": 17.55887802306097, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.427985668182373, "target_z": -0.74999989530677, "ce": 6.308549880981445, "genuineness": 4.168440818786621, "identity": 0.5305273532867432, "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/vn_STNC_reg_low/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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "7a6d5b1611a0b8064918e265a88ef4cbab45e77d89f6c1edcb4543398dc2acff", "generation_seconds_batch": 17.42142223601695, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.418320655822754, "target_z": 0.1367186796592361, "ce": 6.268836498260498, "genuineness": 4.482243061065674, "identity": 0.5302878618240356, "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/vn_STNC_reg_low/1d829402ab2176d2024ae3e918286a39ec094d9a5299caa0e048ba37912e1579.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": "1e3bf3e5bd21a338a2798860c76738f5385a266976bee600448de043598c0eae", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "7e8eedd1515afc262747aa4eefb05b0664b1cda676047c6c79583bceacd55d63", "generation_seconds_batch": 16.388806584989652, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1328877210617065, "target_z": 1.265625045803288, "ce": 5.5659098625183105, "genuineness": 4.846253395080566, "identity": 0.655393660068512, "hypothesis": "Les clés sont sur la table de la cuisine, j'ai regardé dans le tiroir, puis près de la fenêtre. Bon les quand tu partiras.", "wer": 0.041666666666666664, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/1e3bf3e5bd21a338a2798860c76738f5385a266976bee600448de043598c0eae.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "78ecacdc3476d741d8ac9edb0217348ee0a7f5528cb3bc8677e3eeb822f57eab", "generation_seconds_batch": 17.55887802306097, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1233069896698, "target_z": -0.4824217972979934, "ce": 6.165581703186035, "genuineness": 4.362998008728027, "identity": 0.6099697351455688, "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/vn_STNC_reg_low/1eaf0e0e2c966186cf6ada2de41819a315f8b1b1eeeeef5a84a28dd582998db8.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "1eb4f0251017e8b42546b0b90eef7d8a4fea0439e4af0e4206afb6f037f30969", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f4f69776403ca40d3069e14604a84e9053e3b9262cca0ca1758f39d746994a3a", "generation_seconds_batch": 22.871394653921016, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8356503248214722, "target_z": 0.6484375163583171, "ce": 6.125868320465088, "genuineness": 4.519899368286133, "identity": 0.36410561203956604, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/1eb4f0251017e8b42546b0b90eef7d8a4fea0439e4af0e4206afb6f037f30969.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "1f9a141a5c11e4e62c6af85b991ac79ee90ca0f1bb0a0567d6d1739cd84f7a01", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c6faf54e7eb4f47ea6756cb79107e7e93aa5d375f84f097bd6fbb490b9eff079", "generation_seconds_batch": 23.895294410991482, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9246076345443726, "target_z": 0.5703124967283366, "ce": 6.125868320465088, "genuineness": 4.632867813110352, "identity": 0.2508263885974884, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/1f9a141a5c11e4e62c6af85b991ac79ee90ca0f1bb0a0567d6d1739cd84f7a01.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "1fa2239e29c92c401e49a733b1a85d0629a5bef0a3c4571a6cc33adbd84d91f3", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d378a71c3b104a98b97b64e363893af14124c2afc5e6255ff14cd0159b43365d", "generation_seconds_batch": 16.797203914029524, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8979203701019287, "target_z": 0.5937500654332687, "ce": 6.0305562019348145, "genuineness": 4.457139015197754, "identity": 0.54134202003479, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/1fa2239e29c92c401e49a733b1a85d0629a5bef0a3c4571a6cc33adbd84d91f3.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "1fc7fb3067595d7721b8634f7c89286abd254c706882231771bc1396b9253c96", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "17a0af86fdb87d376232eb94458b907ad084ffd6bb917234ff7bc1dbc32c6f8f", "generation_seconds_batch": 17.37296420300845, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6933186054229736, "target_z": 0.7734375687049322, "ce": 5.752562999725342, "genuineness": 4.645420074462891, "identity": 0.7049556970596313, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/1fc7fb3067595d7721b8634f7c89286abd254c706882231771bc1396b9253c96.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "1fdaa85d7f9da1ceacffc19e06eb04679fa809c1344e9cbb573197b4703c3755", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e53d12de27f04ab9c0a74920bd65ee4e9ffb5d68b51629266384a621c2194efd", "generation_seconds_batch": 22.434688107925467, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.022460699081421, "target_z": 0.4843749541967119, "ce": 6.141753673553467, "genuineness": 4.545003414154053, "identity": 0.18392804265022278, "hypothesis": "The 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/vn_STNC_reg_low/1fdaa85d7f9da1ceacffc19e06eb04679fa809c1344e9cbb573197b4703c3755.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "2009bab6dd89e1d745f78168a6c1f5a880f92b728300426f6dcc5a45b5fb4141", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d203e7175e99a540aa2a12eea33d752cd5598fe99dac1081c5f79e956df4bc0c", "generation_seconds_batch": 22.322207071003504, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5623202323913574, "target_z": 0.010253864331812211, "ce": 6.062326908111572, "genuineness": 4.632867813110352, "identity": 0.7619941234588623, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/2009bab6dd89e1d745f78168a6c1f5a880f92b728300426f6dcc5a45b5fb4141.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "203e930b8143dd5d0dba42d835ec3c5b0fa62d6f50a2dc1b2a3579dbfd67b310", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d54e089416a168de8ea624e2c2f45fd8a8b3deed9b8d6407598d93050988e660", "generation_seconds_batch": 23.050241842051037, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0202367305755615, "target_z": 0.4863281110954304, "ce": 5.927301406860352, "genuineness": 4.281409740447998, "identity": 0.41861748695373535, "hypothesis": "The 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/vn_STNC_reg_low/203e930b8143dd5d0dba42d835ec3c5b0fa62d6f50a2dc1b2a3579dbfd67b310.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "207165308733be486874f77710d04b3ce80753b10a70a8eb4fcf35143b782e25", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "964da860464bca3f5710f991934e70b659367b7512d6b1bd97ac3114bd1b35d4", "generation_seconds_batch": 21.786103532998823, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.264869213104248, "target_z": 0.2714844118062137, "ce": 6.3641486167907715, "genuineness": 4.482243061065674, "identity": 0.26582998037338257, "hypothesis": "The 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/vn_STNC_reg_low/207165308733be486874f77710d04b3ce80753b10a70a8eb4fcf35143b782e25.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "20ace774ccbd1d178e40a1de2a41594f5ed296deba970d2942dab9cb88233466", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "101d0fb298662b720d05018453ad9e82fbf3ed3cc7b69c23bcd1974ba6dda2ec", "generation_seconds_batch": 16.957680628984235, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3819681406021118, "target_z": 1.0468750327166343, "ce": 5.982900142669678, "genuineness": 4.65797233581543, "identity": 0.6684248447418213, "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/vn_STNC_reg_low/20ace774ccbd1d178e40a1de2a41594f5ed296deba970d2942dab9cb88233466.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "211b2ad3f5af62812b3d42ae32ebae001e6e12f6b5a17e9fd00d06f3a1a8b6f8", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9de72372b331f3745ff4598f41494afcb6e6f945a97eb06f71edab5dd90444f8", "generation_seconds_batch": 16.9289269100409, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.147000789642334, "target_z": 0.375000052346615, "ce": 5.700935363769531, "genuineness": 4.532451152801514, "identity": 0.625283420085907, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/211b2ad3f5af62812b3d42ae32ebae001e6e12f6b5a17e9fd00d06f3a1a8b6f8.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "22061e6e2b8b4786d30578a1198d721202d8c491b036401c0b583445daea35b5", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c448a41ebd81eed7ddec79f668c34b40d155f73b0cf39348ab813bcc8897aa2f", "generation_seconds_batch": 23.105471969000064, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0936264991760254, "target_z": 0.42187498037001936, "ce": 6.0146708488464355, "genuineness": 4.444586753845215, "identity": 0.4866054952144623, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/22061e6e2b8b4786d30578a1198d721202d8c491b036401c0b583445daea35b5.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "22179882a6136f702fcb8ca2b5c6a0102f1a110611f517f606380c8002560832", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "601e13bec2c918ddcdc987db0a133864e8ed1cae2401eccfcd109e4569440e2a", "generation_seconds_batch": 22.680006399052218, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3819681406021118, "target_z": 1.0468750327166343, "ce": 5.95112943649292, "genuineness": 4.758388519287109, "identity": 0.37231510877609253, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/22179882a6136f702fcb8ca2b5c6a0102f1a110611f517f606380c8002560832.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "223006d54f0506c993e30dff26929df56989fd449672c640835a63171d36b01e", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e34e91c7e93b63de5746d2f485a9042c63c44742dd44366cdffe93c37680389a", "generation_seconds_batch": 18.00824969203677, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9193903207778931, "target_z": 1.4531249672833657, "ce": 5.605623245239258, "genuineness": 4.6705241203308105, "identity": 0.6807188987731934, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/223006d54f0506c993e30dff26929df56989fd449672c640835a63171d36b01e.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": "22aae652fc0d47ec671f2e10c1761f307d75e933c4454d3818f62ca0a7e609b0", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e7adee9fb1a94a0443bb5185da810368dbfb68ea55b4c8421eae597f8624123f", "generation_seconds_batch": 20.415171385044232, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4531340599060059, "target_z": 0.9843749541967118, "ce": 5.863759994506836, "genuineness": 4.507347106933594, "identity": 0.2447989583015442, "hypothesis": "De weg gaat linksaf naar de brug, loopt verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.09523809523809523, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/22aae652fc0d47ec671f2e10c1761f307d75e933c4454d3818f62ca0a7e609b0.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "22b8b657fc3768f01a46904a621e6af731593e4593db2082acb40bef42d35ea2", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4c6d347912dd28f04ad4ac27aee62d1812a1e4e7f4e3c2450d5f7f9f75ab8144", "generation_seconds_batch": 17.41999003395904, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.221845030784607, "target_z": 1.1875000261733075, "ce": 5.152891159057617, "genuineness": 4.469690799713135, "identity": 0.540123462677002, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/22b8b657fc3768f01a46904a621e6af731593e4593db2082acb40bef42d35ea2.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "22f016c7ca03a101f2bb3670fe1398d2a0c5dc04115640d5a74895e68a31da0f", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "99d91d5113e4e94793de11f85a7d4081ebd24713f09de0c0e97b2676becc6186", "generation_seconds_batch": 22.142887281021103, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8742265701293945, "target_z": -0.26367178421133963, "ce": 6.4515180587768555, "genuineness": 4.419482707977295, "identity": 0.6364640593528748, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/22f016c7ca03a101f2bb3670fe1398d2a0c5dc04115640d5a74895e68a31da0f.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "23057beba918a806701aef0da37f262d720518dbfa390041b47f0f78d498a66f", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d73b720c233b8fe4c95e3549affc85e481af2706fea2493b2da7aa596e2b940b", "generation_seconds_batch": 17.184933942044154, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3924028873443604, "target_z": -0.7187500130866538, "ce": 6.372091293334961, "genuineness": 4.256305694580078, "identity": 0.5901362895965576, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/23057beba918a806701aef0da37f262d720518dbfa390041b47f0f78d498a66f.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "234493bdf48fd13a39f0e02733c5c62bc14ac9ac5ae9ca6c9ec7e96ed6fcd7b9", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "de64a6d7e489581d21ada97d42725eab3e1961475562fc01584c0a122a415564", "generation_seconds_batch": 16.515220660017803, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1247615814208984, "target_z": 0.39453120256088015, "ce": 6.062326908111572, "genuineness": 4.532451152801514, "identity": 0.6709911823272705, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/234493bdf48fd13a39f0e02733c5c62bc14ac9ac5ae9ca6c9ec7e96ed6fcd7b9.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "236aa06f43b57967a92c4561326369bb4cd09d0507ec853682cce8bfd20e8027", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5f739f62fab884c0a29f57c2fb4870896e10f269ee4bb4a958edc5ed946bd54e", "generation_seconds_batch": 23.625562317087315, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3160197734832764, "target_z": 0.2265624312950678, "ce": 6.308549880981445, "genuineness": 4.582659721374512, "identity": 0.5353304147720337, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/236aa06f43b57967a92c4561326369bb4cd09d0507ec853682cce8bfd20e8027.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "240cca81fdeeb01a620d3d2ccbb8b2b9bbb325e726ef91ce7af243528366f769", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "265d2cde0b772099d339dd5d255bff2a878b39702428cfdfe5a0fa4c6daba4ff", "generation_seconds_batch": 21.380424850969575, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6933186054229736, "target_z": 0.7734375687049322, "ce": 5.780362129211426, "genuineness": 4.419482707977295, "identity": 0.3341017961502075, "hypothesis": "The 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/vn_STNC_reg_low/240cca81fdeeb01a620d3d2ccbb8b2b9bbb325e726ef91ce7af243528366f769.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "24650576f9bc3e003e649f4242cd35bbcb3cbe019281612851ffa713d6be2d8a", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "435eb970da31475b22b955189127ca58ae8fcd2408f8e549d63571c622d810bf", "generation_seconds_batch": 22.96280747302808, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.831202507019043, "target_z": 0.6523437254625242, "ce": 5.982900142669678, "genuineness": 4.821148872375488, "identity": 0.6636732220649719, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/24650576f9bc3e003e649f4242cd35bbcb3cbe019281612851ffa713d6be2d8a.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "2497bc06e96d18a57c9e9e6b331e0cdea6d1f5e58861358a2b3cc9703d2e266e", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e1d05b3635b9767f283c2eeb00cc740f432111b8ae3ef50a3485a95fd8d2d04e", "generation_seconds_batch": 23.014958665007725, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8534417152404785, "target_z": 0.6328125752482591, "ce": 6.141753673553467, "genuineness": 4.607763767242432, "identity": 0.25513699650764465, "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/vn_STNC_reg_low/2497bc06e96d18a57c9e9e6b331e0cdea6d1f5e58861358a2b3cc9703d2e266e.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "24b45aec8192bff39409e00017cb58f3e3bc51de17e3f1099e9add393b286c6d", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a12967c221fdc10f0c06f806d0aecdc61e250d78b9a4b4b19c546bc44f75dfa7", "generation_seconds_batch": 25.939192579942755, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.386073589324951, "target_z": 0.16503903591773458, "ce": 6.09409761428833, "genuineness": 4.632867813110352, "identity": 0.7290438413619995, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/24b45aec8192bff39409e00017cb58f3e3bc51de17e3f1099e9add393b286c6d.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "2509ba1c703babd241a2a7635ff9a4f77beec56302fee3660c22672be4665732", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "641bf4aa7146462a8324e188600ea71c49960ca01f5afee68238bdfdeb59f57f", "generation_seconds_batch": 15.702437822008505, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3463852405548096, "target_z": 1.0781250196299805, "ce": 6.125868320465088, "genuineness": 4.632867813110352, "identity": 0.738042414188385, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/2509ba1c703babd241a2a7635ff9a4f77beec56302fee3660c22672be4665732.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "2585c97dccd261f3f3fbebdc52a129c8b64badb408be15c1bdc15a9060a06162", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "7eb7f7526c8475bfb7e816011cb36e9be9356d8eef7aea9bd7e1869214ac2ee6", "generation_seconds_batch": 23.19354830996599, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.726335287094116, "target_z": -0.13378915369761832, "ce": 6.3641486167907715, "genuineness": 4.519899368286133, "identity": 0.3841642141342163, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/2585c97dccd261f3f3fbebdc52a129c8b64badb408be15c1bdc15a9060a06162.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "25a369418669493cb3e1cb78a9361bd23b048b3adb54e589b2e974575220d8e0", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7ea28639e1823206adfe3d25a41434d41da03e93351f09ec5764a6e36d59523d", "generation_seconds_batch": 22.424719608039595, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.212949275970459, "target_z": 1.1953125490749517, "ce": 5.689021587371826, "genuineness": 4.532451152801514, "identity": 0.707720160484314, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/25a369418669493cb3e1cb78a9361bd23b048b3adb54e589b2e974575220d8e0.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "25cfdf1adbe1d879e9746977a02900b97689647428c7527323ddfb4d25699314", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "40adf1bc37f89c76c5eab60820d8be21a176bf2c55ba8c232206ec81ec474408", "generation_seconds_batch": 22.79148156696465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9104945659637451, "target_z": 1.4609374901850096, "ce": 5.847874641418457, "genuineness": 4.482243061065674, "identity": 0.3428998291492462, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/25cfdf1adbe1d879e9746977a02900b97689647428c7527323ddfb4d25699314.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "25f9791ef159e052dc0de28973c629fe3ac5e76d9ccd83bb4222fd48b2b4cfa5", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "58a9e34555c65ab95f4ab49b0f356266a5a27fe6e47725ad57190f6dd4da7505", "generation_seconds_batch": 23.625562317087315, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4861505031585693, "target_z": 0.07714844118062136, "ce": 6.229123115539551, "genuineness": 4.331617832183838, "identity": 0.7409506440162659, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/25f9791ef159e052dc0de28973c629fe3ac5e76d9ccd83bb4222fd48b2b4cfa5.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "8a64e8a99db1c582e49d65965064288863d0cdbac615f7c8f9a8e6148c8e31b2", "generation_seconds_batch": 16.299843924934976, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5792081356048584, "target_z": -0.004577606788141523, "ce": 6.316492557525635, "genuineness": 4.356721878051758, "identity": 0.2202284336090088, "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/vn_STNC_reg_low/26d12995a5b024b42cbe037fdee48ffb67b79df656c09f059592185aa881aca9.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "2756bd1d4b8cecc214795942dcbaf5883de0955dcd11e83ce3a50f37ff68fab2", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "051a7156ad295e2ecc55787d1026ecaafe91d70544a1a371dcd8606345097895", "generation_seconds_batch": 16.130435936967842, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.942399024963379, "target_z": 0.5546875556182784, "ce": 5.959072113037109, "genuineness": 4.143336772918701, "identity": 0.36856406927108765, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/2756bd1d4b8cecc214795942dcbaf5883de0955dcd11e83ce3a50f37ff68fab2.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "276060daa82a01e71618c4ecee196f5d6bdb4bec7570955ecb2f66819ce8eade", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c8e3cb11713b562bf722aedc74f8c0b6621e9339eb2d70c1a8fc3786f541a2ef", "generation_seconds_batch": 22.94549871806521, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1951578855514526, "target_z": 1.2109374901850096, "ce": 5.903473377227783, "genuineness": 4.494794845581055, "identity": 0.5565115809440613, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/276060daa82a01e71618c4ecee196f5d6bdb4bec7570955ecb2f66819ce8eade.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": "278eb18592a1869d9d111346c36f63d480cfb6992dd5a6f1a4086d1e87b8be14", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1428c76e51e9f583f8340b7e279f9a15228fa767e05659b43c2abcd3e653bf6d", "generation_seconds_batch": 16.408821127959527, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1684707403182983, "target_z": 1.234374954196712, "ce": 5.5817952156066895, "genuineness": 4.432034492492676, "identity": 0.46305108070373535, "hypothesis": "Ik schreef het nummer op de envelop, lees het goed en controleer de laatste twee cijfers wel 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/vn_STNC_reg_low/278eb18592a1869d9d111346c36f63d480cfb6992dd5a6f1a4086d1e87b8be14.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "27f5b4345f69703286dc57578b6f60b8dffb0b0fbb28c3964a1847b0c6335f22", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "02b3edcd24e26a1896c5a1d2bc17ff15c3dc89a5b14698cc7ccf3ba3d726bd55", "generation_seconds_batch": 23.509212131029926, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9023683071136475, "target_z": 0.5898437516358317, "ce": 5.9431867599487305, "genuineness": 4.174716949462891, "identity": 0.6659961938858032, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/27f5b4345f69703286dc57578b6f60b8dffb0b0fbb28c3964a1847b0c6335f22.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "280c7d49e358e98c79ed3000151deb1fe580ff9cf2128d2771a67dd6550de5d2", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "935ef4916bc4b99549e0f84b1993f1ab0bda2739b523089a0607d835768d75bf", "generation_seconds_batch": 23.089595869998448, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.147000789642334, "target_z": 0.375000052346615, "ce": 5.982900142669678, "genuineness": 4.796044826507568, "identity": 0.7290012836456299, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/280c7d49e358e98c79ed3000151deb1fe580ff9cf2128d2771a67dd6550de5d2.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "288924033352f5f127f390f723542665c9654bf933563fc2b01e20b2f337892f", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "40d6fde90c411724dbbcf5ac5f06387394460622987c1f61fab79641837a1342", "generation_seconds_batch": 23.77284389093984, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6206984519958496, "target_z": -0.04101566671370883, "ce": 6.292664527893066, "genuineness": 4.444586753845215, "identity": 0.17603246867656708, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/288924033352f5f127f390f723542665c9654bf933563fc2b01e20b2f337892f.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "2895465d51ec4be5771f16238bf7a6b2c20a9db1ce54c03af22bb48a1c7e02cc", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "265ee3e63d0190202c0df26356a351386671883da6922b8ce01668b48baa9cbf", "generation_seconds_batch": 22.499867534963414, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.30045223236084, "target_z": 0.24023432019963742, "ce": 5.784333229064941, "genuineness": 4.319066047668457, "identity": 0.5117858052253723, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/2895465d51ec4be5771f16238bf7a6b2c20a9db1ce54c03af22bb48a1c7e02cc.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "28d8afd26a65a40e4067428cffebe69a12a7ab11c79166cfce11db7ffd1092f6", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2ebd00281ac90f1ec41927b8313f590a25a76f72edc08741e7b969502f018308", "generation_seconds_batch": 24.93546195095405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.831202507019043, "target_z": 0.6523437254625242, "ce": 5.8875885009765625, "genuineness": 4.545003414154053, "identity": 0.6049549579620361, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/28d8afd26a65a40e4067428cffebe69a12a7ab11c79166cfce11db7ffd1092f6.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "28f2f9a02005aa00abbe692e94c28add1e92876b8e388a91992c0a9d6bff3e84", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f57a188d000179729c3479da53d117338c1f874c667fdc06279bb0746e97d956", "generation_seconds_batch": 22.747887191944756, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0973048210144043, "target_z": 1.2968750327166343, "ce": 6.125868320465088, "genuineness": 4.984326362609863, "identity": 0.8117275238037109, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/28f2f9a02005aa00abbe692e94c28add1e92876b8e388a91992c0a9d6bff3e84.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": "2923c652e4320ab5febb3dceb845b03f2d51d6fe9f445ff315e0a93ba0a98fef", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a428cedd3d2474c317b0f7d362de89bafc6050d2cb93710119a1adb016fba3d8", "generation_seconds_batch": 16.388806584989652, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.43534255027771, "target_z": 1.0, "ce": 5.363371849060059, "genuineness": 4.7332844734191895, "identity": 0.6491024494171143, "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/vn_STNC_reg_low/2923c652e4320ab5febb3dceb845b03f2d51d6fe9f445ff315e0a93ba0a98fef.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "297bb307b52aa5b7030fbae5ea5eb41f11b5a6e09f54600b498a99f6ea1d2985", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4b5f68da16f122709647508841d6fb67411106c0c5afff7f2d88ece204108140", "generation_seconds_batch": 23.10092718701344, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1247615814208984, "target_z": 0.39453120256088015, "ce": 6.284721851348877, "genuineness": 4.545003414154053, "identity": 0.1257321685552597, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/297bb307b52aa5b7030fbae5ea5eb41f11b5a6e09f54600b498a99f6ea1d2985.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "29e00bcc248cdda22dca22088ec5c00f1886de133e360b8f94e3ceffc3702323", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1f05cfe154b1fda93370d000caab80fc20a5a4b7682a8361c53e190c78c7fc1e", "generation_seconds_batch": 23.34825802803971, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9015988111495972, "target_z": 1.4687500130866538, "ce": 4.848091125488281, "genuineness": 4.494794845581055, "identity": 0.6375756859779358, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/29e00bcc248cdda22dca22088ec5c00f1886de133e360b8f94e3ceffc3702323.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "2a4c2c34241c76f29753abc129e4d395f4e8b442a762c71df62b89d99493e644", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "fa596efffe08aaf116385f0cc6a3934bf0bc6b7161ea91542a14445a91420047", "generation_seconds_batch": 22.147765316069126, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4976125955581665, "target_z": 0.9453125490749515, "ce": 5.2918877601623535, "genuineness": 4.262581825256348, "identity": 0.533665657043457, "hypothesis": "The door's still open. I put your coat on the chair. Tell me when you're ready.", "wer": 0.2222222222222222, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/2a4c2c34241c76f29753abc129e4d395f4e8b442a762c71df62b89d99493e644.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "2aa504a595ac710c5d2e859a8fa25a8a8d093e55caa02bcd222825703160456a", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8557da7a07cd8b192ebae9c7da5694af298f103b94cd8129813651f89070a518", "generation_seconds_batch": 17.066898660967126, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.122537612915039, "target_z": 0.3964843594595987, "ce": 6.23706579208374, "genuineness": 4.7458367347717285, "identity": 0.6190322637557983, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/2aa504a595ac710c5d2e859a8fa25a8a8d093e55caa02bcd222825703160456a.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "2ac592f575af3c65b1215cd35811d1109e865672ef8ae07c825b01f445a4a48a", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "fb7578949f39d870fc2d8e3ad5576944ea5e545cb4b25c863f161d9ac808174d", "generation_seconds_batch": 22.543397480039857, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5865700244903564, "target_z": 0.8671874247517409, "ce": 6.070269584655762, "genuineness": 4.507347106933594, "identity": 0.6357312202453613, "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/vn_STNC_reg_low/2ac592f575af3c65b1215cd35811d1109e865672ef8ae07c825b01f445a4a48a.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "2ae9aa6cd9bbe9c40c01faedddab3d7334e2d9fde8834f930318a501207e8d64", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b830fa1cf591d0e701c36cc6343a44f33cc36726583ce873d5752420a1e251af", "generation_seconds_batch": 23.26596711901948, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2048230171203613, "target_z": 0.3242188105257736, "ce": 6.046441555023193, "genuineness": 4.6956281661987305, "identity": 0.6688041090965271, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/2ae9aa6cd9bbe9c40c01faedddab3d7334e2d9fde8834f930318a501207e8d64.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "2aeb8c1fdb3c66ba11767473881e4d43531770025ed0584cc0f366c4a6695f41", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "475bf8defa1a06d53bda77cd904ee44988733dfa88e86bc053cc83fcfdbadebd", "generation_seconds_batch": 22.682953875046223, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0246846675872803, "target_z": 0.4824217972979934, "ce": 6.006728172302246, "genuineness": 4.457139015197754, "identity": 0.7436274886131287, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/2aeb8c1fdb3c66ba11767473881e4d43531770025ed0584cc0f366c4a6695f41.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "2b2db5c3a7d62ddc461dd8bc311547cb9629c17ed5ed477893545056f14e4163", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a2a41fe3ea209b37d0135278ea7f33c0a275f4d17617ebb256a6e77787b05b96", "generation_seconds_batch": 22.862096181022935, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.70589280128479, "target_z": 1.6406249934566732, "ce": 5.4666266441345215, "genuineness": 4.720732688903809, "identity": 0.6440408229827881, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/2b2db5c3a7d62ddc461dd8bc311547cb9629c17ed5ed477893545056f14e4163.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": "2be3ca52b82b144844182934ee0d0a73cd58681a55be3c52b46f346e846237b1", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a2184522e44b7693d8cbd17ea99c8b5cef37cebeed7eaf9f98c4b357b7195843", "generation_seconds_batch": 20.618699147948064, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5242998600006104, "target_z": 0.9218749803700194, "ce": 5.831989288330078, "genuineness": 4.557555198669434, "identity": 0.3901471495628357, "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/vn_STNC_reg_low/2be3ca52b82b144844182934ee0d0a73cd58681a55be3c52b46f346e846237b1.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "2bef289fd1971a932c709a730756fafae3ff1815cdb39aff689f6e55275cbd18", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7a2b40568bb4b4056789adeb2420fdd6098bdcb2b46ba28bc62fa5d89bc34f2a", "generation_seconds_batch": 16.88415942597203, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3749539852142334, "target_z": 0.17480461102486716, "ce": 5.7684478759765625, "genuineness": 4.507347106933594, "identity": 0.6443461775779724, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/2bef289fd1971a932c709a730756fafae3ff1815cdb39aff689f6e55275cbd18.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "2c1e87ab55c93ba7c99b2f955ce56a74125cb106ae000ab85b25e9f1b77c92e1", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1ab9520fad9e6986a0c44800eb00edd9398812acff372bb66508bd43721d0e97", "generation_seconds_batch": 16.557510982034728, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4086554050445557, "target_z": 1.023437464011702, "ce": 5.700935363769531, "genuineness": 4.7081804275512695, "identity": 0.7996848225593567, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/2c1e87ab55c93ba7c99b2f955ce56a74125cb106ae000ab85b25e9f1b77c92e1.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "2c5ff7361b8540a9eb2764f50b27277b124e062e34c0f43f5afe20a26c428a7c", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6f7f019036adce22d264c9f65d726a3b69f7b8d4e32ce4e0f8efa84d6e992027", "generation_seconds_batch": 23.03135964903049, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3504905700683594, "target_z": 0.19628912752431082, "ce": 5.919358730316162, "genuineness": 4.7332844734191895, "identity": 0.721543550491333, "hypothesis": "He encontrado la cajita, estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/2c5ff7361b8540a9eb2764f50b27277b124e062e34c0f43f5afe20a26c428a7c.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": "2c6c8d34797f3abdd5d694d8bbdcad92ab924b8e3bd1eb344c734898645b5301", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0dea602d910f6fa4ca5da48e162a93a5824eeb5ef0201a6a3b8de1b4be6aeb32", "generation_seconds_batch": 16.30163617595099, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7867238521575928, "target_z": 0.6914062352775145, "ce": 6.046441555023193, "genuineness": 4.507347106933594, "identity": 0.42995429039001465, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het naar de lunch.", "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/vn_STNC_reg_low/2c6c8d34797f3abdd5d694d8bbdcad92ab924b8e3bd1eb344c734898645b5301.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "2c7c99746c205cc776ef395288cf1d6d793d28fb9b2cc8e702016078c97b62fd", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7e10700bb83eb4eda84317abcdab418773b6e0cdda607d8b5f6223d29a9233da", "generation_seconds_batch": 21.852705170982517, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7592672109603882, "target_z": 1.5937499607400387, "ce": 5.5539960861206055, "genuineness": 4.7081804275512695, "identity": 0.6786787509918213, "hypothesis": "Une porte est La porte est encore ouverte J'ai posé ton manteau sur sur la chaise, dis-moi quand tu es prêt.", "wer": 0.2222222222222222, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/2c7c99746c205cc776ef395288cf1d6d793d28fb9b2cc8e702016078c97b62fd.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "2c8dfab06ec6d7880ac8ff203e78514515bff06739aa70b9915c57f8b3b1d3f3", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7bc7d29c257fb98dcd9fa590e4bdb56ba5a087ae655d51636e621b2d83a93994", "generation_seconds_batch": 16.53948803897947, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0624914169311523, "target_z": 0.4492187581791586, "ce": 6.1020402908325195, "genuineness": 4.632867813110352, "identity": 0.6795721054077148, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/2c8dfab06ec6d7880ac8ff203e78514515bff06739aa70b9915c57f8b3b1d3f3.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "2ca45619e2dc64eafca97d6e35319f385f7d1535c6cbb4a70527af343eeaa86a", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "243f82f02a49e3bad0f6a90d957899a169ab2ae8ee3c2bf929eac78613d391e5", "generation_seconds_batch": 16.7476477350574, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7956194877624512, "target_z": 0.6835938170691005, "ce": 5.911416053771973, "genuineness": 4.607763767242432, "identity": 0.618130087852478, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/2ca45619e2dc64eafca97d6e35319f385f7d1535c6cbb4a70527af343eeaa86a.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "2cbedf1d5986e3cc72b1a4ecbb9a9aa9276946094100b549a49b67b16662e818", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aae413c429ed6d400893cb91d67881b450ad51f178fe9e56e6fdc0dc46279a13", "generation_seconds_batch": 22.986409360077232, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3730725049972534, "target_z": 1.0546874509250483, "ce": 5.855817794799805, "genuineness": 4.758388519287109, "identity": 0.8072184920310974, "hypothesis": "J'ai trouvé la la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/2cbedf1d5986e3cc72b1a4ecbb9a9aa9276946094100b549a49b67b16662e818.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": "2cd26d6d76cf3a669a44c615a7104fce197ffb8d711b10a1c95eb391b2cbfde1", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Escribí el número en el sobre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "638093d4bbb9656d649a7b64239f59c67c3e3e142677bb0fcea922f2641bed41", "generation_seconds_batch": 16.13305504701566, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.782275915145874, "target_z": 0.6953125490749515, "ce": 5.561938762664795, "genuineness": 4.582659721374512, "identity": 0.5934708118438721, "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": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/2cd26d6d76cf3a669a44c615a7104fce197ffb8d711b10a1c95eb391b2cbfde1.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "2cda3ad703cf26f79706d3c2c7d4bdf79288e8f63801932cb3b34adc38b20187", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "424c22d49d87a9757ef07c8c5a2a572660249ba3ed855bc9b2106d048979fa48", "generation_seconds_batch": 21.43839480006136, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8660159111022949, "target_z": 1.5, "ce": 5.95112943649292, "genuineness": 4.6203155517578125, "identity": 0.5012091994285583, "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/vn_STNC_reg_low/2cda3ad703cf26f79706d3c2c7d4bdf79288e8f63801932cb3b34adc38b20187.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "2cfa4038ff6e5d9d9a4eb82678544f65bc646e79ee78d8e6fa75a3b845b9fe3f", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0adeef5af0020e6fdfaf06960b0a6fa2edbcb9db3969520a2f83ccc2c2b077d2", "generation_seconds_batch": 21.874275838956237, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.055819511413574, "target_z": 0.45507822887531413, "ce": 6.09409761428833, "genuineness": 4.557555198669434, "identity": 0.4754270315170288, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/2cfa4038ff6e5d9d9a4eb82678544f65bc646e79ee78d8e6fa75a3b845b9fe3f.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "2d85d107f86fa1d38fdc3093f9f8a8db1ca0589693f43685ecd3f0fc68340c03", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "78896dd46f4599d254878ecd7398169ffbcf10048edbe200b05e7d800951815e", "generation_seconds_batch": 22.48926167807076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3160197734832764, "target_z": 0.2265624312950678, "ce": 6.252951145172119, "genuineness": 4.432034492492676, "identity": 0.3526351749897003, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/2d85d107f86fa1d38fdc3093f9f8a8db1ca0589693f43685ecd3f0fc68340c03.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "2dbc1a04f3d81d7e1e416aafdf61e76343a682bdb377ef58dd8e582ce85ef758", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aab73f5de596b63dbbbae03a21f1f3178aa4834f00cf905d978234ee42b27659", "generation_seconds_batch": 18.043714763014577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3538265228271484, "target_z": 0.19335939217623305, "ce": 6.1894097328186035, "genuineness": 4.7081804275512695, "identity": 0.40282100439071655, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/2dbc1a04f3d81d7e1e416aafdf61e76343a682bdb377ef58dd8e582ce85ef758.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "2e46b5a9f70ec6df03c3160bdc89e323dfaf18b4272fbb701592358a5d533a1f", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2eca2a639b7692450c0d4e3ad7642209fd000862a18a75b9f13d4a47dcfd3ada", "generation_seconds_batch": 16.301957082003355, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9727646112442017, "target_z": 1.4062500392599613, "ce": 5.5539960861206055, "genuineness": 4.758388519287109, "identity": 0.7870659828186035, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/2e46b5a9f70ec6df03c3160bdc89e323dfaf18b4272fbb701592358a5d533a1f.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "2e9e8daca3192478b8375396e4503d880643637e964c584a9ec3ac834743d9e5", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d6ca027d8945ba6c728a3b43fa8a0bab114220ba3083eee522ec46cc835ba42d", "generation_seconds_batch": 22.68292859394569, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5109562873840332, "target_z": 0.9335937123758705, "ce": 5.645336627960205, "genuineness": 4.7081804275512695, "identity": 0.694818377494812, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/2e9e8daca3192478b8375396e4503d880643637e964c584a9ec3ac834743d9e5.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "2ea55c329b21faeb82cb6ee997cf8525578bc3a9e9676eec9c3740319cc061e8", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "3e33afd6726b132d95b1e0e9286d13b41e3933c1cb131e47d5416c96b39de805", "generation_seconds_batch": 26.00119538209401, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9104945659637451, "target_z": 1.4609374901850096, "ce": 5.609594821929932, "genuineness": 4.6956281661987305, "identity": 0.2944784462451935, "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/vn_STNC_reg_low/2ea55c329b21faeb82cb6ee997cf8525578bc3a9e9676eec9c3740319cc061e8.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "2ee1cd9a78ad9899aa362994f0f397cfb5005db513a68b3733b90fde45ba05d3", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "54e4a532233a99bcdb4feb0cba0be0f5ae3d0a5ad4c9ca790ff5c9e102f3271c", "generation_seconds_batch": 21.62774611194618, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3782896995544434, "target_z": 0.17187508506324936, "ce": 6.268836498260498, "genuineness": 4.143336772918701, "identity": 0.320873498916626, "hypothesis": "The 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/vn_STNC_reg_low/2ee1cd9a78ad9899aa362994f0f397cfb5005db513a68b3733b90fde45ba05d3.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "2f5936924e0fe0d4e738ca053830de12441db91bc121e8147ee666a083292df0", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f673336eb96e7a723fe398dba531a786b6a36f0b716dea9c24f64739d87f3d94", "generation_seconds_batch": 16.598701850976795, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0624914169311523, "target_z": 0.4492187581791586, "ce": 5.470598220825195, "genuineness": 4.243753433227539, "identity": 0.6727150678634644, "hypothesis": "La puerta sigue abierta. he dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/2f5936924e0fe0d4e738ca053830de12441db91bc121e8147ee666a083292df0.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "2ff07070d2b800c020f34e070847ecf8251ab13a35cbe60b6d744615b1628676", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ec0152903c659ec856e7d3a93355a2a7907efc10fe3302df5f6180aa7b975732", "generation_seconds_batch": 24.293172053061426, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.07061767578125, "target_z": 1.3203124967283366, "ce": 5.40705680847168, "genuineness": 4.7834930419921875, "identity": 0.6563388109207153, "hypothesis": "Me ha encontrado la cajita, estaba junto a la ventana vieja, podemos mirar juntos lo que hay dentro.", "wer": 0.11764705882352941, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/2ff07070d2b800c020f34e070847ecf8251ab13a35cbe60b6d744615b1628676.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "304b7f2f620c3ff656c067f0a74dce12b617ea49b3ea3ff07873bc7e56a9caa8", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "76d39614fc07f6a2dd4222c27c532c21e4ca1bbacb06302c341594709eb9e016", "generation_seconds_batch": 16.408854873036034, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.791171669960022, "target_z": 0.6875000261733075, "ce": 6.046441555023193, "genuineness": 4.758388519287109, "identity": 0.7938989400863647, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/304b7f2f620c3ff656c067f0a74dce12b617ea49b3ea3ff07873bc7e56a9caa8.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "30a9f59006fd889c6e3659617ea106c80a5d7d4cdf778a321f9bc3aafc4b80eb", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "32563e8eb88205c3f52c7a1b0526abaae7cbc454d1e99ec097646af6e2d91212", "generation_seconds_batch": 15.868076462997124, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3249154090881348, "target_z": 0.21875001308665376, "ce": 6.141753673553467, "genuineness": 4.507347106933594, "identity": 0.3401869535446167, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/30a9f59006fd889c6e3659617ea106c80a5d7d4cdf778a321f9bc3aafc4b80eb.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "3116a20b81a78d3a5fd50b3d0adc740714db35b197d2a98285f1d6fcaf0c437f", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0a5be818cb8a0a35b81d3a4f391a5f9d084d7fdf88687c1c6415e795f2d6e8e5", "generation_seconds_batch": 22.775521537987515, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5287476778030396, "target_z": 0.9179687712658123, "ce": 5.426913261413574, "genuineness": 4.557555198669434, "identity": 0.6591790914535522, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/3116a20b81a78d3a5fd50b3d0adc740714db35b197d2a98285f1d6fcaf0c437f.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "31c77ab392bc0c48fa34269d7bda66845273264ddc70f4d8f641c62980f84b00", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6862d239d03700b1bcba5999fc125bb0dd494f5ceb23c3ef61d5c98acb1a6e3e", "generation_seconds_batch": 23.392996742972173, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.079513430595398, "target_z": 1.3124999738266925, "ce": 5.855817794799805, "genuineness": 4.319066047668457, "identity": 0.5066881775856018, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/31c77ab392bc0c48fa34269d7bda66845273264ddc70f4d8f641c62980f84b00.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "31dc078f723865acd40783a2e6f0a94800b41d93f7db84a5d1d22fdc4f1a1f5c", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "71a78ff0dd442773c8db849b3d46a1f1210dac3658bcc0237820ed6f495256d8", "generation_seconds_batch": 16.889224926009774, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8445460796356201, "target_z": 0.6406249934566731, "ce": 6.38797664642334, "genuineness": 4.6956281661987305, "identity": 0.48375338315963745, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/31dc078f723865acd40783a2e6f0a94800b41d93f7db84a5d1d22fdc4f1a1f5c.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": "321f2c902e04001a79b912349fccaceb329bfe152d24e061a1d4883c96a473b0", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Escribí el número en el sobre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d277a161aed01068310494fe2972ac5e0116ecab50c7f88e5bff278cb5144bce", "generation_seconds_batch": 20.4698285079794, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7555887699127197, "target_z": 0.7187500130866538, "ce": 5.764476776123047, "genuineness": 4.6705241203308105, "identity": 0.6485748291015625, "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/vn_STNC_reg_low/321f2c902e04001a79b912349fccaceb329bfe152d24e061a1d4883c96a473b0.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "328c3afa10212a6e47863785c2582b787646a5eb287c4000b81922f8dcd45d73", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5aa13aafe0bb729dfec978abfc5011c7041f68e66b07a3bba9e602b0a5aa0202", "generation_seconds_batch": 17.706129434052855, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7046518325805664, "target_z": -0.1147460833215728, "ce": 6.0146708488464355, "genuineness": 4.281409740447998, "identity": 0.4133393168449402, "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/vn_STNC_reg_low/328c3afa10212a6e47863785c2582b787646a5eb287c4000b81922f8dcd45d73.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "3381f2b225bb6faaf5f60a7fc373aa97895aad45cb9b6c31516d305cb160f769", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "12b1c9ca6b5ca76115c2eb612490d3c5b7afcaed49531d57d64aba627f08c9b1", "generation_seconds_batch": 16.768817565985955, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0210061073303223, "target_z": -0.39257804566216165, "ce": 6.173524379730225, "genuineness": 4.024092197418213, "identity": 0.2633449137210846, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/3381f2b225bb6faaf5f60a7fc373aa97895aad45cb9b6c31516d305cb160f769.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "338b4874d9d25cec6894e98dd7134bee14b7e1bd59da0e0c1d3c231742d37301", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f40db5874fc5ad1f2c77cc3723d384664157a50d90cce5971c0c28a82c239dc5", "generation_seconds_batch": 23.757514734053984, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2040536403656006, "target_z": 1.2031249672833657, "ce": 5.4825119972229, "genuineness": 4.68307638168335, "identity": 0.7923243641853333, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/338b4874d9d25cec6894e98dd7134bee14b7e1bd59da0e0c1d3c231742d37301.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "33976377c33d7e8d7ef66c9f8ed64ced0e9955e355618b83c0e1bcb4a9766b05", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b23e395db6c88d6a40098d1e363632e5dd3d7ad7e467ccab9dcd9f8cb8a38e37", "generation_seconds_batch": 16.26477698306553, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9727646112442017, "target_z": 1.4062500392599613, "ce": 5.661221981048584, "genuineness": 4.770940780639648, "identity": 0.5767004489898682, "hypothesis": "He encontrado la cajita, estaba junto a la ventana vieja, podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/33976377c33d7e8d7ef66c9f8ed64ced0e9955e355618b83c0e1bcb4a9766b05.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "d363ba9defbfdedb0aeb3196d338462879821295a51951e7d751b83fe0fb7f8d", "generation_seconds_batch": 17.8559917450184, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.39052152633667, "target_z": 0.16113272212029756, "ce": 6.022613525390625, "genuineness": 4.331617832183838, "identity": 0.7327215671539307, "hypothesis": "De weg gaat linksaf naar de brug. Loopt verder, tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/33f63538ac2c49676cc53462aa2024b47e58cb29b8690774a6580785ecb0e60e.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "345b089547ab7dd6ff61e0568a789e21b0c340b0fe4d9b58cfc8e4f677795121", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "53c2757846ec2f181472d46de238fe8b1ca61822f8f3ef276fc24b217e21d53c", "generation_seconds_batch": 16.901126254932024, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.670736789703369, "target_z": -0.08496085938903544, "ce": 6.24500846862793, "genuineness": 4.6705241203308105, "identity": 0.7243541479110718, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/345b089547ab7dd6ff61e0568a789e21b0c340b0fe4d9b58cfc8e4f677795121.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "dd5ad79a4eba770f373bd5fe7ab838f4eaba7318707b7c851ce3d5755ab1563c", "generation_seconds_batch": 15.173490614979528, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2011446952819824, "target_z": -0.5507812418208414, "ce": 6.467403411865234, "genuineness": 4.168440818786621, "identity": 0.10857188701629639, "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/vn_STNC_reg_low/347b084eb1c87968a3fc14b855043a2de140e81cf0874b7d65525285bfb31414.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "35131334b61151b177c68126082668ae60b9bcd0405eafd8bd4249412c1306e6", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6588dc6829e98c253f46d7d8d0510fa78a135157ad23b091da357e03768fe3a2", "generation_seconds_batch": 21.590833863941953, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.275219440460205, "target_z": 1.1406249934566732, "ce": 6.062326908111572, "genuineness": 4.582659721374512, "identity": 0.412276953458786, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/35131334b61151b177c68126082668ae60b9bcd0405eafd8bd4249412c1306e6.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "356110b307cfe264b2fedf1747f5b736455d67c991aad53f8ba49eddbb5af960", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c7451688a4238f5b1a102c5d31e80c21235f448227193d6bd4cca66173afb5b5", "generation_seconds_batch": 17.313607475021854, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.465579032897949, "target_z": 0.09521493310730762, "ce": 6.316492557525635, "genuineness": 4.570107460021973, "identity": 0.7264031767845154, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/356110b307cfe264b2fedf1747f5b736455d67c991aad53f8ba49eddbb5af960.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "369e79533dc0d010ac975b87d8e1b91bd5783e995cd065b1d678be984f545a90", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bf0cd608739b4ef195d6fa92f137a1741290fe1e3b391d8a27f8f602e7ebb011", "generation_seconds_batch": 22.142331150942482, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.862337589263916, "target_z": 0.624999947653385, "ce": 6.308549880981445, "genuineness": 4.65797233581543, "identity": 0.36227479577064514, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/369e79533dc0d010ac975b87d8e1b91bd5783e995cd065b1d678be984f545a90.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "36e173b19ccca04c0316ec829ccc17964f8adae8f2eddd20b46a456e85bd6e7f", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b744d04176f98dbc1c9db7e2f282f30488928f4b15ee2f2c00b26bd4f4fbf6f9", "generation_seconds_batch": 16.978285963996314, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.837531805038452, "target_z": -0.2314453235418641, "ce": 6.2767791748046875, "genuineness": 4.18099308013916, "identity": 0.36990296840667725, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/36e173b19ccca04c0316ec829ccc17964f8adae8f2eddd20b46a456e85bd6e7f.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "37a0b2e5cc7dac0dd68827cfd9edea2a3e9227153065adc31ee7f50abdd1d930", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "969877aad617dad5b531e82306c52f260a6dd5d98107909064ac4755114a352a", "generation_seconds_batch": 20.804628188023344, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7503714561462402, "target_z": 1.6015624836416829, "ce": 5.2918877601623535, "genuineness": 4.645420074462891, "identity": 0.7469393610954285, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/37a0b2e5cc7dac0dd68827cfd9edea2a3e9227153065adc31ee7f50abdd1d930.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": "38340bbbb7eecee84a43ab7611c807eca0028fad52bf65c53043d75ef17e623f", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ecb1bf032f158e82c04ee49d6a1ed1a394ed24dac3eddbb182708e966089d544", "generation_seconds_batch": 20.164742170018144, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.80006742477417, "target_z": 0.6796875032716634, "ce": 5.808161735534668, "genuineness": 4.7081804275512695, "identity": 0.6533100008964539, "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/vn_STNC_reg_low/38340bbbb7eecee84a43ab7611c807eca0028fad52bf65c53043d75ef17e623f.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": "3844601dd97541397309834ddaf993e889d28300e4697a3eb456dee55948bf4e", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "235c74461fefb5bac553f3d13c72683d61c7981fde9f478285c8644e0ff32583", "generation_seconds_batch": 16.423975200974382, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1862621307373047, "target_z": 1.2187500130866538, "ce": 5.855817794799805, "genuineness": 4.770940780639648, "identity": 0.8396393060684204, "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/vn_STNC_reg_low/3844601dd97541397309834ddaf993e889d28300e4697a3eb456dee55948bf4e.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "384c6aa0249aacf7821b5e6714daa37630c7965a446ed91f591c64872378c49d", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3bb487a0206cf06df9d192df2d463b0ea97c21696a40271d9ea4415f56c9e07d", "generation_seconds_batch": 16.97683608799707, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2381820678710938, "target_z": 0.29492187581791585, "ce": 6.062326908111572, "genuineness": 4.381826400756836, "identity": 0.5479562282562256, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/384c6aa0249aacf7821b5e6714daa37630c7965a446ed91f591c64872378c49d.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "397d00d26e460e6bc0a614594d27d94006fa61ddd20e6ae74c998db777566d3d", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fea0f981bb8fc7b572c8c2cb06f4dfb8fba0ee003c1bd4d28ac0d2e8683283cb", "generation_seconds_batch": 15.779992735944688, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7200058698654175, "target_z": 0.75, "ce": 6.340320587158203, "genuineness": 4.582659721374512, "identity": 0.3284637928009033, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/397d00d26e460e6bc0a614594d27d94006fa61ddd20e6ae74c998db777566d3d.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "39a6068fda3443bce32fa7bfcd2563eab3b3337ab48f282698429db1411d9a08", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "29af2fead0fba20bf1f7df86183deb751e9cd39308a55077cf7f24ff18923179", "generation_seconds_batch": 21.874275838956237, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.657735824584961, "target_z": 0.8046874509250485, "ce": 5.990842819213867, "genuineness": 4.519899368286133, "identity": 0.7172287106513977, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/39a6068fda3443bce32fa7bfcd2563eab3b3337ab48f282698429db1411d9a08.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "39b7cd704f9f0f2585e018a21e9d5da50ff028cb518b6964dd2b04b2272aa424", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "33009d23eee2731cb9a6b2e66a90caf69728694fa9bf88e935f9b7d593f6ac73", "generation_seconds_batch": 22.650859603076242, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.102522134780884, "target_z": 0.4140625621616053, "ce": 6.23706579208374, "genuineness": 4.607763767242432, "identity": 0.5799897909164429, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/39b7cd704f9f0f2585e018a21e9d5da50ff028cb518b6964dd2b04b2272aa424.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "39c5dddcec3c875f3c1f21c9610f70271ff127600239a3b117dd222eb05d1573", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7c246270019e18e97ac072a62b9c16a9f02deef22e6ddbc5082bb8ef88c2af8d", "generation_seconds_batch": 17.22883652697783, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3819681406021118, "target_z": 1.0468750327166343, "ce": 5.49045467376709, "genuineness": 4.758388519287109, "identity": 0.5916613936424255, "hypothesis": "He encontrado la cajita, estaba junto a la ventana vieja, podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/39c5dddcec3c875f3c1f21c9610f70271ff127600239a3b117dd222eb05d1573.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "3a6b02a49c0445326956d98e5bf019478d645db67b39a13847df188c5810bf5a", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a5f52774d768090d719e453254938f42c856b632a0e1a6705b6eb63fae7a9f46", "generation_seconds_batch": 22.85736859298777, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6703099012374878, "target_z": 1.6718749803700195, "ce": 5.95112943649292, "genuineness": 4.6705241203308105, "identity": 0.46138280630111694, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/3a6b02a49c0445326956d98e5bf019478d645db67b39a13847df188c5810bf5a.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "3a7aff32f2c0138eb862d9994e14441b9a6c3c323ba7419b2ed4f7ac32dc5a31", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8151dc4553050cb33f5c302bbf1eeae5bcc1c8e7eba18efccabfc624863dc774", "generation_seconds_batch": 17.41752178000752, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.840867757797241, "target_z": -0.23437505888994187, "ce": 6.157639026641846, "genuineness": 4.557555198669434, "identity": 0.7956446409225464, "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/vn_STNC_reg_low/3a7aff32f2c0138eb862d9994e14441b9a6c3c323ba7419b2ed4f7ac32dc5a31.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "3af1448c5b6e92a631b744c22e10e53df70fc24cffb41d909c2364cb32898d3c", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4957c8f84a51ee5207f963f1b80ab66f4a3a1fdc6e084032cf79603c4218d0f6", "generation_seconds_batch": 17.21068907994777, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1966967582702637, "target_z": -0.5468749280234044, "ce": 5.967014789581299, "genuineness": 4.061748504638672, "identity": 0.44519567489624023, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/3af1448c5b6e92a631b744c22e10e53df70fc24cffb41d909c2364cb32898d3c.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "3b0406c2e19f766dddaf11a96303bde564b5924232dcab5076bd80a469db811a", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f564a7f518d04c73f001bbe0429b8824f0adcb60f0d0f37afe50e89e4ea504cc", "generation_seconds_batch": 21.904287863057107, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3819681406021118, "target_z": 1.0468750327166343, "ce": 6.205295085906982, "genuineness": 4.7081804275512695, "identity": 0.4413158893585205, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/3b0406c2e19f766dddaf11a96303bde564b5924232dcab5076bd80a469db811a.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "3b688c9f5af0e37f6a7c0125c2a8274f1d1833c8852e103de8f171a679873744", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cb8c2b797a8fe7ffffac86e6931c3e8e0cfed5bd5123729d27cb109642a70bb1", "generation_seconds_batch": 23.03135964903049, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.278212785720825, "target_z": 0.2597656798003626, "ce": 5.982900142669678, "genuineness": 4.770940780639648, "identity": 0.7926291823387146, "hypothesis": "He encontrado la cajita, estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/3b688c9f5af0e37f6a7c0125c2a8274f1d1833c8852e103de8f171a679873744.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "3b71447ce46ccb603adfb5bb92de09ba3d14b1c4f30e96b680160b6afe742cdf", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9fee97c13f2aac71cbeeb18605643171277f7521f5cf5425ce25c36d98f167fb", "generation_seconds_batch": 23.19354830996599, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3360350131988525, "target_z": 0.20898443797952115, "ce": 6.23706579208374, "genuineness": 4.469690799713135, "identity": 0.5584168434143066, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/3b71447ce46ccb603adfb5bb92de09ba3d14b1c4f30e96b680160b6afe742cdf.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "3beff5cab1f5045d9dabdb660c1c3c9ce47f30f9f337b810c2cc0b0dc83c9327", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "7163332e656077876da25f4fc35db991537d337c279bb48ee9c2fe072a9f77f9", "generation_seconds_batch": 16.01192609698046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.674072742462158, "target_z": -0.0878905947371132, "ce": 6.260893821716309, "genuineness": 4.406930446624756, "identity": 0.1631564199924469, "hypothesis": "The 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/vn_STNC_reg_low/3beff5cab1f5045d9dabdb660c1c3c9ce47f30f9f337b810c2cc0b0dc83c9327.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "3c1a21bf7cbefa45e5834d9bb71589110d29a357150fd338b091421d4ec936d2", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7a32fa378286063b993ea70e53ac9316dca462ab33946c39653dd7eb35a1a545", "generation_seconds_batch": 22.315808710991405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4842690229415894, "target_z": 0.9570312810808026, "ce": 5.935244083404541, "genuineness": 4.6956281661987305, "identity": 0.7815189361572266, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre nous pouvons regarder ensemble ce qu'elle contient", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/3c1a21bf7cbefa45e5834d9bb71589110d29a357150fd338b091421d4ec936d2.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "3c34f719d8cb4142260b0e8ae31ae88d957ee75a5911c30e4dbd4202571cfa40", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "96cf4c75a8b3c29476e20c4d02cb10ce741f6eb64c1a28049db310f6ab64451c", "generation_seconds_batch": 17.066898660967126, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3671700954437256, "target_z": 0.18164066017038194, "ce": 6.062326908111572, "genuineness": 4.7834930419921875, "identity": 0.7564388513565063, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/3c34f719d8cb4142260b0e8ae31ae88d957ee75a5911c30e4dbd4202571cfa40.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "9f706f86e39ab3b28da09688ad64e9f8032254f54fd7e4292c19cca6b442c3d8", "generation_seconds_batch": 18.105520284036174, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.463568687438965, "target_z": -0.7812499869133462, "ce": 6.149696350097656, "genuineness": 4.369274139404297, "identity": 0.7367116808891296, "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/vn_STNC_reg_low/3c590f9302048ad206a88d3c2440f1028cc9eea1f1662250baf365be21a6ebfa.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "3c92ed074a191f32a835426b7ef843963367ed25f6100c7ca3899dffae806abe", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "61ee099256ba4748dc6beccb42989c8c851586eb3159b22256e471f2ca056108", "generation_seconds_batch": 22.53665775002446, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0091171264648438, "target_z": 0.496093686202563, "ce": 5.95112943649292, "genuineness": 4.1998209953308105, "identity": 0.5748783349990845, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/3c92ed074a191f32a835426b7ef843963367ed25f6100c7ca3899dffae806abe.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "3ca8eca805b0920a6431c052a1f04074bea9abd46c894cfccd22138c745ab49b", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "59cd529af27bd108fd7142c8e48364701e5d7cfc2f026593d17193eb14c33010", "generation_seconds_batch": 21.988879265030846, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0439305305480957, "target_z": 1.3437499607400387, "ce": 5.71682071685791, "genuineness": 4.519899368286133, "identity": 0.4107862114906311, "hypothesis": "The 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/vn_STNC_reg_low/3ca8eca805b0920a6431c052a1f04074bea9abd46c894cfccd22138c745ab49b.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "3cbc3b8d57e46c78a2b7813248497561c4f3a21cea044d87f1a71a616597025f", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "408eefa1aeec237d06f8442967c408b443cd308c4d253d80a91d443038fb7771", "generation_seconds_batch": 22.871394653921016, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6666314601898193, "target_z": 0.7968750327166344, "ce": 5.760505676269531, "genuineness": 4.21237325668335, "identity": 0.43973493576049805, "hypothesis": "The 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/vn_STNC_reg_low/3cbc3b8d57e46c78a2b7813248497561c4f3a21cea044d87f1a71a616597025f.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": "3cf1408e4d303f027a81576127fdbdb62404147c43bceb42ca97edaa86ddab22", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Your message arrived after I had left.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "56cf5af4438ba372013cf95bcb5844b2811bffc0c1f6f5841dadd1dd5d10fd03", "generation_seconds_batch": 17.504247745964676, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5509870052337646, "target_z": 0.8984375163583171, "ce": 5.935244083404541, "genuineness": 4.3127899169921875, "identity": 0.4704659581184387, "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/vn_STNC_reg_low/3cf1408e4d303f027a81576127fdbdb62404147c43bceb42ca97edaa86ddab22.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "3d06248b859976d6d0196f7a9b256c34ef164fc1e25aa27ffc758a30941e39f2", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "69f44c1d015432be1df61b7b4d1a5f9afaec13cd7d9d40af3fd94b149954ca00", "generation_seconds_batch": 22.605406500049867, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.182583808898926, "target_z": 0.3437499607400388, "ce": 5.8875885009765625, "genuineness": 4.243753433227539, "identity": 0.23395219445228577, "hypothesis": "The 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/vn_STNC_reg_low/3d06248b859976d6d0196f7a9b256c34ef164fc1e25aa27ffc758a30941e39f2.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": "3d3b729bc470f640379b43ca31fc5fc4a5c5e856a999253de7fefd6c6e1762a9", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7c3bc2d7ffd8f2e2e609b11de88cb60f0d805da1d424c7b708795cb78a0b573b", "generation_seconds_batch": 20.893695896025747, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5420912504196167, "target_z": 0.9062500392599613, "ce": 5.661221981048584, "genuineness": 4.607763767242432, "identity": 0.7747460603713989, "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/vn_STNC_reg_low/3d3b729bc470f640379b43ca31fc5fc4a5c5e856a999253de7fefd6c6e1762a9.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": "3d75b762a86ed11ac3fc74f6c367b7db1dd6855ff2504d688f722339668d4639", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The keys are on the kitchen table.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "be61ee80b983b237610b41cf704e9d057c04a9fcc4dde9df3569e3e2c365d054", "generation_seconds_batch": 23.769224004936405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.43534255027771, "target_z": 1.0, "ce": 5.935244083404541, "genuineness": 4.532451152801514, "identity": 0.4641103148460388, "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/vn_STNC_reg_low/3d75b762a86ed11ac3fc74f6c367b7db1dd6855ff2504d688f722339668d4639.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "3d7bd025b8068db9536a12b18b9686701e4ff9f9457d99bae2a8f2867f488a0e", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c7ba02815c31477964b2f5d58d202e1cc4c781400ad51a01630dcfd0b7149d18", "generation_seconds_batch": 26.00119538209401, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8838074207305908, "target_z": 1.484374954196712, "ce": 5.8875885009765625, "genuineness": 4.7332844734191895, "identity": 0.3699226975440979, "hypothesis": "I f found the little box. It was beside the old window. We can look inside together.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/3d7bd025b8068db9536a12b18b9686701e4ff9f9457d99bae2a8f2867f488a0e.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "3d7fd3143e019a90b47927b9b464d270bcc7e16bce3afdbf86115e215badb995", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "75cf1a582d85687f7ce1394df656f78ff183c0d67d67b1303ca907f06112a3d5", "generation_seconds_batch": 16.374500519013964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7252233028411865, "target_z": -0.13281257524825907, "ce": 5.998785495758057, "genuineness": 4.406930446624756, "identity": 0.6618672609329224, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/3d7fd3143e019a90b47927b9b464d270bcc7e16bce3afdbf86115e215badb995.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": "3db7a4122f4367f98eee850dec93dc13198375db51784b9d8a4a8ad507f6ea3c", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "64202d6b07bf3aebfabd970421a9f25278d27dfa6d57caba0f909b448ff1dfb2", "generation_seconds_batch": 20.618699147948064, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.457581877708435, "target_z": 0.9804687450925048, "ce": 5.824047088623047, "genuineness": 4.557555198669434, "identity": 0.14380481839179993, "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/vn_STNC_reg_low/3db7a4122f4367f98eee850dec93dc13198375db51784b9d8a4a8ad507f6ea3c.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "3efa0c42dc0ec1b2ed2523d5e06ebead3bf705d6810b52b1dccc7366f1f18443", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1d671cf2c222af0f2c4b612d345c63a39a96c15ee3643ac6c7e1c54b1858aaa7", "generation_seconds_batch": 17.302102377056144, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.251525640487671, "target_z": 0.28320314381206474, "ce": 6.181467056274414, "genuineness": 4.507347106933594, "identity": 0.5114301443099976, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/3efa0c42dc0ec1b2ed2523d5e06ebead3bf705d6810b52b1dccc7366f1f18443.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "3f11186891c20bc4d6446eda5e884b5a8be1d7c41fdcb7697dbf6255858caacc", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d19c907876be4c3cef9c2a0477bf880a2aaa38570fff7f84d823fab28f9494ee", "generation_seconds_batch": 23.951120762038045, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3108023405075073, "target_z": 1.1093750065433268, "ce": 5.6334228515625, "genuineness": 4.934117794036865, "identity": 0.5539884567260742, "hypothesis": "He encontrado la cajita, estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/3f11186891c20bc4d6446eda5e884b5a8be1d7c41fdcb7697dbf6255858caacc.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "3f67ba75cda89595288f300279bba2b6c181482a9d8160a40411abb2587230df", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "912f3eafe6d1e5fcdf97a257cdeeb7ffd72e2cd24156a2e7d29e3f1a4832788d", "generation_seconds_batch": 22.475721878930926, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3093478679656982, "target_z": 0.23242190199122337, "ce": 6.205295085906982, "genuineness": 4.582659721374512, "identity": 0.5701277852058411, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/3f67ba75cda89595288f300279bba2b6c181482a9d8160a40411abb2587230df.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "3f9706e25d2afbc6433e15bf41f1916a6c881cb3bdbf8815c34ea4c7ae66a1c8", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8223e6c9ae72f9abc198ace36a1fc5c81dad153916f52a064b1b257aa93df8d8", "generation_seconds_batch": 17.39467202196829, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.061721920967102, "target_z": 1.3281250196299805, "ce": 5.732706069946289, "genuineness": 4.356721878051758, "identity": 0.6173473596572876, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/3f9706e25d2afbc6433e15bf41f1916a6c881cb3bdbf8815c34ea4c7ae66a1c8.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "3fe6f56d6f5939867415e722f9982bf7bea5a151c0c47998b5421f2c8d79cd09", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b3a031652e6feb0ed016db8017fa154fda8ab64f9ec78aa26b647f8a10f726fe", "generation_seconds_batch": 16.76424115197733, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.39052152633667, "target_z": 0.16113272212029756, "ce": 5.95112943649292, "genuineness": 4.494794845581055, "identity": 0.4815143942832947, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/3fe6f56d6f5939867415e722f9982bf7bea5a151c0c47998b5421f2c8d79cd09.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "400da597362af1be3b2477e1b86ee875558bcccf0de2dfaf0355cf490f9c50eb", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3e073ca036b1acbec9d6bbb8b0ca2fba56c7f8d9d9942a4521e88794e76f2ca1", "generation_seconds_batch": 23.705427067005076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9727646112442017, "target_z": 1.4062500392599613, "ce": 5.4825119972229, "genuineness": 4.7332844734191895, "identity": 0.534475564956665, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/400da597362af1be3b2477e1b86ee875558bcccf0de2dfaf0355cf490f9c50eb.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "40126aa1f9b08687542ea6e2bc40e2904de8d17c1b77c85bc7fd98b8e4d98a55", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9dada3f34929189957d44c1ac80ba4175d50b2f40747e6f91c41ca1ffee4d64f", "generation_seconds_batch": 16.4634020839585, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4338881969451904, "target_z": 0.12304679075466649, "ce": 5.998785495758057, "genuineness": 4.796044826507568, "identity": 0.7842944264411926, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/40126aa1f9b08687542ea6e2bc40e2904de8d17c1b77c85bc7fd98b8e4d98a55.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "40ac0782773bb93cdb1d31e1278c4369d60953961d44ea696e003b72743faf27", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2688aaad5e897a85c61c63db4253b6a8691ae8d6261cc0bbd7533a90e54d253b", "generation_seconds_batch": 22.858420604025014, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.042476177215576, "target_z": 0.46679675149470523, "ce": 5.967014789581299, "genuineness": 4.457139015197754, "identity": 0.7967783808708191, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/40ac0782773bb93cdb1d31e1278c4369d60953961d44ea696e003b72743faf27.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "40c77fcfe994beee58839992324d30eba7cf35525278690b5ca0c715f09dcf3f", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "61a822b29934ad4d89a21b802cfd2144f0340babf9eb132c2b7e5b2945704e47", "generation_seconds_batch": 22.53774720407091, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3856465816497803, "target_z": 1.9218749803700195, "ce": 5.704906940460205, "genuineness": 4.344170093536377, "identity": 0.43909698724746704, "hypothesis": "De deur staat nog open. Ik heb je je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.05263157894736842, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/40c77fcfe994beee58839992324d30eba7cf35525278690b5ca0c715f09dcf3f.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "40d18062501a2658bd9ad203a338de2efa1a4d6f8ee42386d278baefe9b737e9", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a4c07fc4333c2961a42d1b96a9eec4c7e27afc765c53a89987530d0fcb1d3bf2", "generation_seconds_batch": 18.510509132989682, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.834195852279663, "target_z": -0.22851558819378634, "ce": 6.157639026641846, "genuineness": 4.118232727050781, "identity": 0.26028725504875183, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/40d18062501a2658bd9ad203a338de2efa1a4d6f8ee42386d278baefe9b737e9.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "4101a6f328bd9f466fec96d4c7dba31d4f547514bcf2d12b1dc20c7493e446f1", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "efaaa4999be6ccb85927c71866f64f93d7068fa0cee57a8c03065c0e830a82db", "generation_seconds_batch": 17.022994811995886, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0350347757339478, "target_z": 1.3515624836416829, "ce": 5.5738525390625, "genuineness": 4.6705241203308105, "identity": 0.4259564280509949, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/4101a6f328bd9f466fec96d4c7dba31d4f547514bcf2d12b1dc20c7493e446f1.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "41857665ee8d73db473ca54e4b3604166238469f9289255ceea4f8853edf6bab", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f00fe0497b6745d93f9a28ceffdb6c49eac17c1645535519bb34d259339511d1", "generation_seconds_batch": 21.743415127974004, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1328877210617065, "target_z": 1.265625045803288, "ce": 6.038498878479004, "genuineness": 4.507347106933594, "identity": 0.4262188673019409, "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/vn_STNC_reg_low/41857665ee8d73db473ca54e4b3604166238469f9289255ceea4f8853edf6bab.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "4188ba9c28af1d11712e76987af97dcf357f36fb9fcdb192848eba9e0106b39b", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8a1d05bdaddb5335ab9593a9ae3fa4988f83d2e02e96321918a387bb6ed893e8", "generation_seconds_batch": 22.692677541985177, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9601905345916748, "target_z": 0.5390625098149903, "ce": 5.816103935241699, "genuineness": 4.306513786315918, "identity": 0.7320985198020935, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/4188ba9c28af1d11712e76987af97dcf357f36fb9fcdb192848eba9e0106b39b.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "41cb63af96f441f50558955d9131b5877e0ef7246002e2e1162aa2b430e396cc", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "57e1721aa5891a47abfa805bb3a1df95ff0e8fd55796a071f6b18e5cf99e4b96", "generation_seconds_batch": 22.34314983896911, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.456683397293091, "target_z": 0.1030273513157217, "ce": 6.268836498260498, "genuineness": 4.607763767242432, "identity": 0.5612995028495789, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/41cb63af96f441f50558955d9131b5877e0ef7246002e2e1162aa2b430e396cc.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "421c3a758d925f50728fe76a3dee1e06f6604e52dd259908a737d29522db6cba", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "41561df3678229c5d7c4886cc22678e5178d40d9b1001723fea55adfbb15f15e", "generation_seconds_batch": 24.103616160107777, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2040536403656006, "target_z": 1.2031249672833657, "ce": 5.601652145385742, "genuineness": 4.607763767242432, "identity": 0.7100944519042969, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/421c3a758d925f50728fe76a3dee1e06f6604e52dd259908a737d29522db6cba.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": "4257f404d956b511c209f1ca83b7c3d6ddc21bd936f14e5544698292c8088a79", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "b2d284b8c93c15cd4f5d835cc4680222c3a14d8b597ae64be1de01abf18f6fc4", "generation_seconds_batch": 20.49811221694108, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3552809953689575, "target_z": 1.0703124967283366, "ce": 5.7128496170043945, "genuineness": 4.65797233581543, "identity": 0.6756272315979004, "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": 0.120000000000001, "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/vn_STNC_reg_low/4257f404d956b511c209f1ca83b7c3d6ddc21bd936f14e5544698292c8088a79.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "42a7a44686db87629fe5c033f4b8a6aacc1ddb98411c549a7049be7f6d61808c", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "09bfd7f9b65974202d9497a544e9f5a0f8555dd27562d8d4626575a5fef20a56", "generation_seconds_batch": 17.178188227000646, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9053616523742676, "target_z": -0.29101556202047885, "ce": 6.109982967376709, "genuineness": 3.8797430992126465, "identity": 0.41098013520240784, "hypothesis": "The 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/vn_STNC_reg_low/42a7a44686db87629fe5c033f4b8a6aacc1ddb98411c549a7049be7f6d61808c.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "42f1ce0e55482cb064674d89a3b36a8df5e5dd7bd13978c84d1fce56fdb3b810", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "78f284514bf2ab784bc242388ed85d7ef3c3b64fdb694d9e48198a81da81e2d9", "generation_seconds_batch": 17.462195041007362, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.548142671585083, "target_z": 0.02270503017468275, "ce": 6.197352409362793, "genuineness": 4.532451152801514, "identity": 0.48365724086761475, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/42f1ce0e55482cb064674d89a3b36a8df5e5dd7bd13978c84d1fce56fdb3b810.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "5f9b4a25e81b5fbd28ea5e4e095be726a9031e5ef4d8d8f862af9197d5c48af7", "generation_seconds_batch": 17.8559917450184, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6840803623199463, "target_z": -0.09667959139488654, "ce": 6.213237762451172, "genuineness": 4.519899368286133, "identity": 0.8545365333557129, "hypothesis": "De weg gaat linksaf naar de brug. Loopt verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "423cb1e6c35aa6864f67e5e036733e5f7767a2c65f475e7d1a99d73ddb6d55b4", "generation_seconds_batch": 17.193959300988354, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.744126558303833, "target_z": -0.14941399011444645, "ce": 6.054384231567383, "genuineness": 4.607763767242432, "identity": 0.7163881063461304, "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/vn_STNC_reg_low/43d771a5c055e2aea1186dc99f4cdef2b4c6bc8c6b2779ad35a29ca11c75a91c.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "444b614539916887f62aef4d9d5fa72de1b16509373732cc36a64882acc64a50", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "827e69aebeeac03b5dcb52814593d87048523fb23a72efd70f0b4f663ae59493", "generation_seconds_batch": 17.5902086350834, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5450847148895264, "target_z": 0.025390620910420703, "ce": 6.054384231567383, "genuineness": 4.557555198669434, "identity": 0.4770188331604004, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/444b614539916887f62aef4d9d5fa72de1b16509373732cc36a64882acc64a50.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "4499c0811f5ba659c6ca27d2592520004223b4f41157ca23e9469cf0ab4ac921", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e78b09abbfb4c85f246f76f899210170e12ea2a4ac46bedaa3c63fe1b572567b", "generation_seconds_batch": 16.797203914029524, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.195927381515503, "target_z": 0.33203122873418767, "ce": 6.205295085906982, "genuineness": 4.570107460021973, "identity": 0.596057653427124, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/4499c0811f5ba659c6ca27d2592520004223b4f41157ca23e9469cf0ab4ac921.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "44dc4760785d0ad8573544fdc4d560fc26132d3a76a1bea873b166573f6576d4", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4bdf9f0a573c04fb09b0a3c43fa853a62415a11c0aaca9cc4fd979df16884679", "generation_seconds_batch": 22.711376299965195, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.122537612915039, "target_z": 0.3964843594595987, "ce": 5.959072113037109, "genuineness": 4.362998008728027, "identity": 0.5795947313308716, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/44dc4760785d0ad8573544fdc4d560fc26132d3a76a1bea873b166573f6576d4.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "450378b25991f2ce51f8dc05762f5fed43ed42ca7ba7271ff64f1d756168f6d5", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "44a51c5dcddc4c114ebf7f0e3642ec819e907db042349432b31ba08a135d76be", "generation_seconds_batch": 15.668553957017139, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8178589344024658, "target_z": 0.6640624574683753, "ce": 6.062326908111572, "genuineness": 4.8588056564331055, "identity": 0.6878349781036377, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/450378b25991f2ce51f8dc05762f5fed43ed42ca7ba7271ff64f1d756168f6d5.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "456b0006507d4be81d3994c4209e12be398c7f4bff1c0a74406048f7928d779d", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "05ad7d2a1c594f640d4507976b6966d6434394674256d4c03009d2b86742b7a4", "generation_seconds_batch": 23.84609296103008, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3552809953689575, "target_z": 1.0703124967283366, "ce": 5.748591423034668, "genuineness": 4.7332844734191895, "identity": 0.6151883602142334, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/456b0006507d4be81d3994c4209e12be398c7f4bff1c0a74406048f7928d779d.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "456bb2a6b3989375721155ec4f98d780587699960bcb1e4f02ba61ce3393bcc3", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5908b72eae25d9453941e814713eb29fc459002209fcc44ae04a43f0481eda52", "generation_seconds_batch": 24.79618810804095, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.221845030784607, "target_z": 1.1875000261733075, "ce": 5.967014789581299, "genuineness": 4.381826400756836, "identity": 0.5651103258132935, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/456bb2a6b3989375721155ec4f98d780587699960bcb1e4f02ba61ce3393bcc3.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "45f75dac07ae6d19cba16665ac5fbe894c7d39c7768aba9a8d2e4421b7b256df", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9d2447d18bf5a3c4af52b907e99d927c9af4eabe23042d77dc9b78dde15adb53", "generation_seconds_batch": 23.199607295915484, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3108023405075073, "target_z": 1.1093750065433268, "ce": 5.764476776123047, "genuineness": 4.532451152801514, "identity": 0.6791131496429443, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist,", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/45f75dac07ae6d19cba16665ac5fbe894c7d39c7768aba9a8d2e4421b7b256df.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "39df0cabec75b3d69aa939d4b61636fa2d978bb8ee2a608c94e98dc520c59035", "generation_seconds_batch": 15.965999093023129, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.459120750427246, "target_z": -0.7773436731159092, "ce": 6.348263263702393, "genuineness": 4.193545341491699, "identity": 0.6108206510543823, "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/vn_STNC_reg_low/45f9ec8497c8691f9e18d794cf98f47c522bfab041b36df5accabf5c5c456412.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "4627795ef45534f7dcf7647bf7b7a61fc21ee827ccda50b81279c81f75d6f7de", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "14905143693473d9adfe46c967abe69e021ca1a46fbb154fb7f6cfccec5777f2", "generation_seconds_batch": 16.08077994803898, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7111101150512695, "target_z": 0.757812522901644, "ce": 5.7247633934021, "genuineness": 4.218649387359619, "identity": 0.4982438385486603, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/4627795ef45534f7dcf7647bf7b7a61fc21ee827ccda50b81279c81f75d6f7de.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "4628157e2ad73a5a3b8c61d152ff29e775c950cd709ba8faa01a2ead2a1d8576", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "717c229dc3e76e36b2ba4c51e418f9d87fab16fd897fa0c50e60919e0c8a9bf5", "generation_seconds_batch": 21.62774611194618, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6774086952209473, "target_z": -0.09082033008519097, "ce": 6.070269584655762, "genuineness": 4.049196243286133, "identity": 0.33368775248527527, "hypothesis": "The 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/vn_STNC_reg_low/4628157e2ad73a5a3b8c61d152ff29e775c950cd709ba8faa01a2ead2a1d8576.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "4659e04208c240b87bdedf09b94b7e1fa6841cbe0e2e11e6c0ea74f244a68648", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ef0899a4eee4663676878114be51a9bf93d33ec453f9c5190e4364bdc3a98f53", "generation_seconds_batch": 16.149424509028904, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.695200204849243, "target_z": -0.10644537588847909, "ce": 5.9431867599487305, "genuineness": 4.293961524963379, "identity": 0.6094188690185547, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/4659e04208c240b87bdedf09b94b7e1fa6841cbe0e2e11e6c0ea74f244a68648.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "46f0337577aa2c8d469ee9fdda7891ca20eff116aaa026785cf33c67d3be8f0e", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cc4cf0746c6877a44bb504bd60177bf23cc96d7369aae805af815800b5857ce6", "generation_seconds_batch": 22.420944545068778, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8578896522521973, "target_z": 0.6289062614508221, "ce": 6.181467056274414, "genuineness": 4.645420074462891, "identity": 0.7487814426422119, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/46f0337577aa2c8d469ee9fdda7891ca20eff116aaa026785cf33c67d3be8f0e.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "478fb1578c0c3e8a589c499d08e982d5bf83e6533838f1aff8fa82d2ef49a769", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "95902a085c5dab06899426a4f696c7d6c2ad2cc3dfb36c62bc461b86d96168b7", "generation_seconds_batch": 25.338029153994285, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2396365404129028, "target_z": 1.1718749803700195, "ce": 5.895530700683594, "genuineness": 4.482243061065674, "identity": 0.4325484037399292, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/478fb1578c0c3e8a589c499d08e982d5bf83e6533838f1aff8fa82d2ef49a769.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "47c59171140ea9d209578a86f1ee797ad267decc48ca968b170fc663bbc82d6f", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "84a83280c7ec32e8a64dd9891f75370bedb336b67549c60de4fc03808a15c663", "generation_seconds_batch": 17.65638886601664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7244536876678467, "target_z": 0.746093790895793, "ce": 6.125868320465088, "genuineness": 4.425758361816406, "identity": 0.605891764163971, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/47c59171140ea9d209578a86f1ee797ad267decc48ca968b170fc663bbc82d6f.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "6f79ba7692c9a33c904357f4b9c2a25f437fb3d9ba6d7411bf1bf5655ea1209a", "generation_seconds_batch": 16.350541777093895, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1692402362823486, "target_z": 0.3554686927458898, "ce": 5.998785495758057, "genuineness": 4.7834930419921875, "identity": 0.7386715412139893, "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/vn_STNC_reg_low/47d00b404f769e82194cfb39f2be7a6bd3cc5bfaac4f38356b0fada2abdb9f4d.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "48603139ec3fa345ab3f0c89c3316343b5dbb899a01fa8f8eca158d718483591", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0ce50318cae28899b2caf4f33accb1841cd18c67a20f15c452fbe47e6e5f53ca", "generation_seconds_batch": 23.432092441013083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8393287658691406, "target_z": 1.523437464011702, "ce": 5.506340026855469, "genuineness": 4.7332844734191895, "identity": 0.6827960014343262, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/48603139ec3fa345ab3f0c89c3316343b5dbb899a01fa8f8eca158d718483591.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "48f047b759846dd4ae6a67a2e335d9a231815d36d125dbdebfe820090f0dfcb9", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "73b285255c00a3747e95d16717065fa65422dcd88a044bb28037f83340e6623b", "generation_seconds_batch": 18.615822600084357, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7641420364379883, "target_z": -0.16699219281645308, "ce": 6.1894097328186035, "genuineness": 4.607763767242432, "identity": 0.34662243723869324, "hypothesis": "De deur 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/vn_STNC_reg_low/48f047b759846dd4ae6a67a2e335d9a231815d36d125dbdebfe820090f0dfcb9.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "499314a1aea79123e88fb6594c4743152c6176bc63b4362d4647f3f880b9a732", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "83a13eee98242c2681d26cc4358f232109eb4ccf3e8aa33299cd0540e7588818", "generation_seconds_batch": 26.506137130898423, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4483437538146973, "target_z": 0.11035148029945614, "ce": 6.308549880981445, "genuineness": 4.432034492492676, "identity": 0.6655510663986206, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/499314a1aea79123e88fb6594c4743152c6176bc63b4362d4647f3f880b9a732.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "11b4c8633c27e6d89c0464fd7d236aaf3843675c0f3a54df5927c89d1d684b0e", "generation_seconds_batch": 17.819578128983267, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1522183418273926, "target_z": -0.507812627594874, "ce": 6.086154937744141, "genuineness": 4.268857479095459, "identity": 0.691536545753479, "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/vn_STNC_reg_low/49d468ae6540cc7c1e5b0e6883c25e2b29599458c9de0a5398000eaca62677ab.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "4a06ebf2aa11d338d91b2a40bba32504324f494f211b007a2d4c43b22248a046", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "05f44d19a48c24ce0fb089d6c2d854dfe2cdbb55b1b24821f8d32bdefb477858", "generation_seconds_batch": 16.169855204992928, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6221528053283691, "target_z": 0.8359375425316247, "ce": 5.824047088623047, "genuineness": 4.65797233581543, "identity": 0.5946922898292542, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/4a06ebf2aa11d338d91b2a40bba32504324f494f211b007a2d4c43b22248a046.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "4a7214bc7b8cc97c3fb79252360acdc5cf1a1bfca058e3d8405c0ead94b03819", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3641a923acb6d210396c5c4746c2df8f4f8e54b40200e0052591bfdaadab6213", "generation_seconds_batch": 16.618353783967905, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.640713691711426, "target_z": -0.05859366002925548, "ce": 5.967014789581299, "genuineness": 4.457139015197754, "identity": 0.7793179750442505, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/4a7214bc7b8cc97c3fb79252360acdc5cf1a1bfca058e3d8405c0ead94b03819.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "4a8d428faef33c9f3efad949670f2b59f131bc52400a4216dcbc77c5b53b4565", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8c819caee07e364a1ed3742eb770da0c04e56bd3bb4c91efeea5e68c7ce3bffb", "generation_seconds_batch": 22.29878579999786, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5065083503723145, "target_z": 0.9375000261733075, "ce": 6.09409761428833, "genuineness": 4.632867813110352, "identity": 0.4728391766548157, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/4a8d428faef33c9f3efad949670f2b59f131bc52400a4216dcbc77c5b53b4565.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "4aff4c6a46b56e004c18048fa556435292b44bfbfcb34186e8c7526e50d1937e", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "79dc777c16a56e48f3ee5992ad5c9f433202a67ab953b46113cc976dae7ff411", "generation_seconds_batch": 16.40111226902809, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3374894857406616, "target_z": 1.0859375425316247, "ce": 5.629451274871826, "genuineness": 4.375550270080566, "identity": 0.5758565068244934, "hypothesis": "If the door is still open, I put your coat on the chair. Tell me when you are ready.", "wer": 0.05555555555555555, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/4aff4c6a46b56e004c18048fa556435292b44bfbfcb34186e8c7526e50d1937e.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "4b4731e15c91281804e00b15b19399e759ef4414a1ea7c0d5457e8d38ad2bb6b", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "06beeb46f5fc21cc7a0749698c0a5e4a771dca45001d65137a6172b920912a3f", "generation_seconds_batch": 23.719138300977647, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.102522134780884, "target_z": 0.4140625621616053, "ce": 6.09409761428833, "genuineness": 4.482243061065674, "identity": 0.7504132986068726, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/4b4731e15c91281804e00b15b19399e759ef4414a1ea7c0d5457e8d38ad2bb6b.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "4bf7f8c75775293c5b79030bdd7401dde572e989439522b136bdffcd538ec199", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5bd8fe34fa0dba1e8c8887f007a7c2f223127bb1d0833cb92be4ec3c412eff01", "generation_seconds_batch": 15.668553957017139, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8631069660186768, "target_z": -0.253906209104207, "ce": 5.959072113037109, "genuineness": 4.344170093536377, "identity": 0.8613554239273071, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la ver Biaxel, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.3684210526315789, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/4bf7f8c75775293c5b79030bdd7401dde572e989439522b136bdffcd538ec199.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "4c0318db94f3af42326acfc634604ef409446102e0a1fbe6ea0333cffd4efcd7", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7c1548f4faf77e3caa2b68622a40444e44b86fd49364046257daa18c9fd58c35", "generation_seconds_batch": 25.40135297004599, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0002212524414062, "target_z": 0.503906313797437, "ce": 6.24500846862793, "genuineness": 4.770940780639648, "identity": 0.7819137573242188, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/4c0318db94f3af42326acfc634604ef409446102e0a1fbe6ea0333cffd4efcd7.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "4c3ee3f981c364eb462888a428637c43267bb64172e9b5299b73f13fc41b9c1c", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "241909c80211d7d6b00888f386c33e17026e79fe168069de4d79b27f13c9451a", "generation_seconds_batch": 17.20774588396307, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.986877679824829, "target_z": 0.5156250458032882, "ce": 6.109982967376709, "genuineness": 4.519899368286133, "identity": 0.3561052680015564, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/4c3ee3f981c364eb462888a428637c43267bb64172e9b5299b73f13fc41b9c1c.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": "4c5684f19a71019a685fad55c8bdbd2ae494cb61509ed3e16465259fea343b04", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a00246d5f9f75593378c77fdeb4feb20a00d1b8f29fadd5fbf67e6f51e539b73", "generation_seconds_batch": 15.662657730979845, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.488716959953308, "target_z": 0.9531249672833656, "ce": 5.6254801750183105, "genuineness": 4.720732688903809, "identity": 0.6374490857124329, "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/vn_STNC_reg_low/4c5684f19a71019a685fad55c8bdbd2ae494cb61509ed3e16465259fea343b04.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "4c5c965d0463fc225c6d61a1dbc9e2f3127295c8001d3a1beced3f00734b2ad5", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f7c4248023f19e355c3126bcf71c913d4380a3cb848bdf383abfe71514f2e590", "generation_seconds_batch": 16.35425183805637, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8853464126586914, "target_z": -0.2734375687049322, "ce": 6.23706579208374, "genuineness": 4.369274139404297, "identity": 0.2396744042634964, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/4c5c965d0463fc225c6d61a1dbc9e2f3127295c8001d3a1beced3f00734b2ad5.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "4cd956048190b4df1dbd846246669f5bf4f178319f753c10f58931e0f84a6c77", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eb88993597d3004cad9a3f62a75a9d217a76338fd3e07e23019f71d59fd3be45", "generation_seconds_batch": 16.38734126102645, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.791171669960022, "target_z": 0.6875000261733075, "ce": 5.974957466125488, "genuineness": 4.218649387359619, "identity": 0.6075749397277832, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/4cd956048190b4df1dbd846246669f5bf4f178319f753c10f58931e0f84a6c77.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "4cfed5a379c7ff9c859290d53ebbe27b9eb5691e5129c2ff4b11a6c807fc5f2b", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ce465dc31c289f3417b613e654fc67ba8abd5f653f85665491b8ff5b7e591b21", "generation_seconds_batch": 22.749273675959557, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4709254503250122, "target_z": 0.9687500130866538, "ce": 5.609594821929932, "genuineness": 4.250029563903809, "identity": 0.45936113595962524, "hypothesis": "The 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/vn_STNC_reg_low/4cfed5a379c7ff9c859290d53ebbe27b9eb5691e5129c2ff4b11a6c807fc5f2b.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "4d0226e4cd25cbc63045ebc7e72402e48ab022c162ca365f4971330e9a067ae5", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ad0d3ee301858191724851c2fe66ddbd063fba42fe8f8e23210e14be4fbf6fc4", "generation_seconds_batch": 16.618353783967905, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3293633460998535, "target_z": 0.21484369928921673, "ce": 6.117925643920898, "genuineness": 4.770940780639648, "identity": 0.7366957664489746, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/4d0226e4cd25cbc63045ebc7e72402e48ab022c162ca365f4971330e9a067ae5.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "4d0ccafb993d999201f735d7eba225533a57934132e574d4ea3a611109ec7955", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9a9b0405dcffdc1419e57d3c9c55f26d4346afe13701c46a7234b7058e3078bc", "generation_seconds_batch": 21.696319633047096, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4798212051391602, "target_z": 0.9609374901850097, "ce": 5.2918877601623535, "genuineness": 4.7332844734191895, "identity": 0.5887231826782227, "hypothesis": "La puerta sigue abierta. he dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/4d0ccafb993d999201f735d7eba225533a57934132e574d4ea3a611109ec7955.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "4dffcdc28748b941b7d6e3d936421a29e367f3b2cd1e60b52cf788374b6cbd82", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5c3b8521418deb330e1ef5943529c6313e84046b1b08a69310ca38d8324fab23", "generation_seconds_batch": 17.96961906296201, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9364967346191406, "target_z": -0.31835933982961806, "ce": 5.8875885009765625, "genuineness": 4.306513786315918, "identity": 0.6634825468063354, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/4dffcdc28748b941b7d6e3d936421a29e367f3b2cd1e60b52cf788374b6cbd82.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "4e135f7e2cd49b1635852dda75b81d0c4bf7b6ad83ff64bf35e0a6c7f04bcdb7", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1c3b20df48c5d65375316880ba54211493ae93b31525a8f96da88438c8b100eb", "generation_seconds_batch": 22.598050444037654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1417834758758545, "target_z": 1.2578125229016441, "ce": 5.617537498474121, "genuineness": 4.8588056564331055, "identity": 0.6589000225067139, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/4e135f7e2cd49b1635852dda75b81d0c4bf7b6ad83ff64bf35e0a6c7f04bcdb7.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "4e4a7e54bdf0de1ecd810bc345dc58619f7c3819154c46b85f22af4c51a2bb89", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ad3d17e651d6a03ceb388b5b70e0aae3b20e8fe1bf944eb743999e5a2eec1f1b", "generation_seconds_batch": 22.290384648018517, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8786745071411133, "target_z": -0.26757809800877663, "ce": 5.77639102935791, "genuineness": 3.9676074981689453, "identity": 0.566740095615387, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/4e4a7e54bdf0de1ecd810bc345dc58619f7c3819154c46b85f22af4c51a2bb89.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "4e607ab9a3dd30df619460c08e788e3cc190a59a01613199cdf37e9fe3e27bbb", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "755f4e8e08769a3d4a050b91ac4d13f881590ce78e923c4c6fe0c319cbbfd974", "generation_seconds_batch": 16.97683608799707, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7111101150512695, "target_z": 0.757812522901644, "ce": 6.23706579208374, "genuineness": 4.595211505889893, "identity": 0.6879156827926636, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/4e607ab9a3dd30df619460c08e788e3cc190a59a01613199cdf37e9fe3e27bbb.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": "4e89bf7ea5a0c1e6a5e74d187d0ae637edaf3143fd15de143e38c5630d53476a", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1e63f34569395611e318963ac1cecf1e321557020272925984801a6f8e4fcacd", "generation_seconds_batch": 20.41448323999066, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3819681406021118, "target_z": 1.0468750327166343, "ce": 5.982900142669678, "genuineness": 4.519899368286133, "identity": 0.6321778297424316, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten naar de lunch terug.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/4e89bf7ea5a0c1e6a5e74d187d0ae637edaf3143fd15de143e38c5630d53476a.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "4f12451bac85191fd0d4f8b9d09889a6c2901dc8fc4d779a801ccfca240e4fff", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c5bfe5136d84378cec2472912f460f68f13cb3a220d7b59d0b204f60c2af43c8", "generation_seconds_batch": 22.80731120100245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.102522134780884, "target_z": 0.4140625621616053, "ce": 6.09409761428833, "genuineness": 4.306513786315918, "identity": 0.6496400833129883, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/4f12451bac85191fd0d4f8b9d09889a6c2901dc8fc4d779a801ccfca240e4fff.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "4f3adb9755b8ad4f1a1aa53d024ec7af8dfa07d08493ade7d38d593d262b651d", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a64e708179d1d4321f16d1d75cc8c6cedec1a999950c78e3cef3ea1fab8f22b3", "generation_seconds_batch": 16.34151714190375, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2559735774993896, "target_z": 0.27929683001462774, "ce": 6.300607204437256, "genuineness": 4.65797233581543, "identity": 0.634217381477356, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/4f3adb9755b8ad4f1a1aa53d024ec7af8dfa07d08493ade7d38d593d262b651d.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "4f663fb3c99a0e22c0ccd90c41a2f2940acf1c66f00cbf3bea9c915cf12644aa", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "00d6f858bbeca392021ac5faba6439a51528ea55515bbb3d02c2c95bab37398b", "generation_seconds_batch": 22.89835396001581, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9601905345916748, "target_z": 0.5390625098149903, "ce": 6.252951145172119, "genuineness": 4.419482707977295, "identity": 0.13497956097126007, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/4f663fb3c99a0e22c0ccd90c41a2f2940acf1c66f00cbf3bea9c915cf12644aa.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "4f79f51007c46a6dae309e4fbe0ebcc59bbba2d9984554e70a402ae7edb54abb", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c5f237fa1afe2530860e0fe31b037f1f506550f56fae377c44efc5eadab00e22", "generation_seconds_batch": 24.498039223020896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9549732208251953, "target_z": 1.4218749803700195, "ce": 5.689021587371826, "genuineness": 4.432034492492676, "identity": 0.6369744539260864, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/4f79f51007c46a6dae309e4fbe0ebcc59bbba2d9984554e70a402ae7edb54abb.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "4f7f5a54a2b6c9bc6f9fe197a464f76bb77c4c41c81778b3806b663ae8093ddc", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "16a2853358fcae50529f2944252151f22cb7c7be6c74a267e23cd05bf0b4282d", "generation_seconds_batch": 16.80170062603429, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1002984046936035, "target_z": 0.41601550967386386, "ce": 5.998785495758057, "genuineness": 4.457139015197754, "identity": 0.4482996463775635, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/4f7f5a54a2b6c9bc6f9fe197a464f76bb77c4c41c81778b3806b663ae8093ddc.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "4faa60ad2eec2a0893c7145f0cbde6c36acd269a4140094dcc6ed525f41de851", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "64ddcaaf373d3280086ca6c52c99a3e8f9d92e338ce1db2678839b742c2e2629", "generation_seconds_batch": 21.099137362092733, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.126985549926758, "target_z": 0.39257804566216165, "ce": 5.895530700683594, "genuineness": 4.6956281661987305, "identity": 0.5970514416694641, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/4faa60ad2eec2a0893c7145f0cbde6c36acd269a4140094dcc6ed525f41de851.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "5037cc79903be0d09b866b776ca9bffdaa418c56ed140ef6b5f3b776b57476fd", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "de532d8f6e0dab1c17b133e8a1ca5fa8d0853b82d17e1eef5e93c8e2981b2e71", "generation_seconds_batch": 22.986359643982723, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0232300758361816, "target_z": -0.39453120256088015, "ce": 6.324435234069824, "genuineness": 4.469690799713135, "identity": 0.4574458599090576, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/5037cc79903be0d09b866b776ca9bffdaa418c56ed140ef6b5f3b776b57476fd.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "503f5781776656d9b5a67e03e8905ba2f272c89cc49dee62a865d0115feeade8", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "15eed786ff5b0564c43ae35c70aa784f00ed0c4a60165516590f71a772a822fb", "generation_seconds_batch": 25.40135297004599, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9157118797302246, "target_z": 0.5781250196299806, "ce": 6.086154937744141, "genuineness": 4.808597087860107, "identity": 0.7996801137924194, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/503f5781776656d9b5a67e03e8905ba2f272c89cc49dee62a865d0115feeade8.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "508f7f26b4c5242d9d6c20b1a68d53661f49d611ca67ed19654b07af5644429e", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e3f5d638c736fe5b88f9d4b7c715233b4dc09a8f02bf5896e8ebf6f0f6ed8eb4", "generation_seconds_batch": 23.400134368101135, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.751140832901001, "target_z": 0.7226563268840908, "ce": 6.229123115539551, "genuineness": 4.632867813110352, "identity": 0.66629958152771, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/508f7f26b4c5242d9d6c20b1a68d53661f49d611ca67ed19654b07af5644429e.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": "50aa08e7936e81279a3afda9aa3e7049dbb327a62caf5a74ccf39993cead6e2c", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ba85064b328f6419ffd21eb4b00671916d90166b1da0625112da7b639b69bd15", "generation_seconds_batch": 20.242094643996097, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7333494424819946, "target_z": 0.738281267994149, "ce": 5.8875885009765625, "genuineness": 4.632867813110352, "identity": 0.7264045476913452, "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/vn_STNC_reg_low/50aa08e7936e81279a3afda9aa3e7049dbb327a62caf5a74ccf39993cead6e2c.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": "50c479d8ad9e5a94b80d7cfebe9dc8e6e997146a4fbacc9a2eb79dfb4302ae3e", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e74d4e8be53cd2b94805fad9e845016839f6d37b9ec4109550dd75f45182a12b", "generation_seconds_batch": 17.604474638937972, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7644845247268677, "target_z": 0.7109374901850097, "ce": 5.784333229064941, "genuineness": 4.846253395080566, "identity": 0.6471875905990601, "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/vn_STNC_reg_low/50c479d8ad9e5a94b80d7cfebe9dc8e6e997146a4fbacc9a2eb79dfb4302ae3e.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "50faaddf260ee6cd8e436c86e48d6d219b450ad2d03d24b989e7a2f82b195aac", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8ed205dd869152d76ca8f55bee89f5620bf3bea3a4060ddfc0356a425f7a447d", "generation_seconds_batch": 16.88433393498417, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9164814949035645, "target_z": -0.3007813465140714, "ce": 6.427689552307129, "genuineness": 4.344170093536377, "identity": 0.5620147585868835, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/50faaddf260ee6cd8e436c86e48d6d219b450ad2d03d24b989e7a2f82b195aac.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "50fe48565c9ff990c60a78eb067ad4db328b21a858575ccc8c1116a48c9d7498", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "584e0aa051beb1831e4023971a122315159ce9907224b635261aeb694a3dfe59", "generation_seconds_batch": 21.380424850969575, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.488716959953308, "target_z": 0.9531249672833656, "ce": 6.006728172302246, "genuineness": 4.2751336097717285, "identity": 0.6131390333175659, "hypothesis": "The door is still open. I I put your coat on the chair, t tell me when you are ready.", "wer": 0.1111111111111111, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/50fe48565c9ff990c60a78eb067ad4db328b21a858575ccc8c1116a48c9d7498.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "511033a775b0812f9f5d84497cdd5ba0aa4fe6004fc3092ea6db2fba6e3203f7", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ea62ea661aff9a1236165a9dc6b3fcdaa72d9d79789b7a6dfb04d455e602755c", "generation_seconds_batch": 23.432092441013083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8037458658218384, "target_z": 1.5546874509250483, "ce": 5.5659098625183105, "genuineness": 4.7834930419921875, "identity": 0.571231484413147, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/511033a775b0812f9f5d84497cdd5ba0aa4fe6004fc3092ea6db2fba6e3203f7.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "514a0c810ced2605984b4241644c4f57c36818fffdbd1e99adfcebb7f8f75779", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2730ac3df13cc2b16ec465557a7e5d7835d52c11dd4c6eae540734ef4bbdff86", "generation_seconds_batch": 24.24937932402827, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5154041051864624, "target_z": 0.9296875032716634, "ce": 5.9431867599487305, "genuineness": 4.808597087860107, "identity": 0.7351933717727661, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/514a0c810ced2605984b4241644c4f57c36818fffdbd1e99adfcebb7f8f75779.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "514b81b5c690028b3aae9476629e75302ae14518c6e386a7992fd0ea2897946c", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "11cf76f96c719040a6491426d4ce22764dec5350d3b172a31a167debde50234a", "generation_seconds_batch": 22.43666469398886, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.657735824584961, "target_z": 0.8046874509250485, "ce": 5.7247633934021, "genuineness": 4.607763767242432, "identity": 0.5955847501754761, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/514b81b5c690028b3aae9476629e75302ae14518c6e386a7992fd0ea2897946c.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": "51942b134e273284468ca6c9d40966a41ddfd96e8edd16204d9d8dc11e9f632f", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5e9e55815a85d870f1691a36410d6b82303a14be19a69c3c13dbd00418036e63", "generation_seconds_batch": 20.173583226976916, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1328877210617065, "target_z": 1.265625045803288, "ce": 5.4388275146484375, "genuineness": 4.582659721374512, "identity": 0.7939823865890503, "hypothesis": "Mon message est arrivé après mon départ. Je l'ai relu. Puis j'ai arrangé mon téléphone. Dis-moi quand nous pourrons parler.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/51942b134e273284468ca6c9d40966a41ddfd96e8edd16204d9d8dc11e9f632f.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "51a229844a778e2e0823c913160e19284f06b37fb569e5cbde4a8948df2d4d16", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "89e00daa088ea62d9a052c3c62b4c9666966901bd9ecca1d834c854d6a7e028e", "generation_seconds_batch": 24.79618810804095, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9379512071609497, "target_z": 0.5585937647224855, "ce": 5.990842819213867, "genuineness": 4.595211505889893, "identity": 0.7623691558837891, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/51a229844a778e2e0823c913160e19284f06b37fb569e5cbde4a8948df2d4d16.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "51acf4e4ec81b071ca8dcb96e50cf97d229caa3c313e2d44ce88f8b33aa681e7", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "8736c5e1b411d67c26b00f40b37f950a7ff12099387ce523901e50ab5caf17f3", "generation_seconds_batch": 21.988879265030846, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4709254503250122, "target_z": 0.9687500130866538, "ce": 5.5817952156066895, "genuineness": 4.174716949462891, "identity": 0.5667871236801147, "hypothesis": "The 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/vn_STNC_reg_low/51acf4e4ec81b071ca8dcb96e50cf97d229caa3c313e2d44ce88f8b33aa681e7.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": "51fd3c9fab6153a95b808efe5fa2ac1f199c22fd01e19b269f7370f3e911d36d", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The road turns left behind the bridge.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1b7e6b8ea0495f4592db0f4b46b99d40da17d57a19b0def3ed5ec95993f8104e", "generation_seconds_batch": 22.58502525393851, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8756811618804932, "target_z": 0.6132812156475339, "ce": 5.764476776123047, "genuineness": 4.281409740447998, "identity": 0.374936044216156, "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/vn_STNC_reg_low/51fd3c9fab6153a95b808efe5fa2ac1f199c22fd01e19b269f7370f3e911d36d.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "520bc4047e9ea572f6bbc3388e0185ec172cd87352690b84450788fd36db3cfd", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cf29088af6e8b1f6cdff6f9947efa702b3c639c75094bb739f4371d76c5668d8", "generation_seconds_batch": 20.91106655402109, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.293010950088501, "target_z": 1.124999947653385, "ce": 5.879645347595215, "genuineness": 4.545003414154053, "identity": 0.5013422966003418, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/520bc4047e9ea572f6bbc3388e0185ec172cd87352690b84450788fd36db3cfd.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "5261a38e5343da72ccad504ab1d39e1611077814b2042a1a3cda5c1632300ad8", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "71862dd0b8a2686115487d63dff4f5f28e5d8870c7552cd626b1008fba8bfe54", "generation_seconds_batch": 15.702437822008505, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.178135871887207, "target_z": 0.3476562745374758, "ce": 6.09409761428833, "genuineness": 4.557555198669434, "identity": 0.5378468036651611, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/5261a38e5343da72ccad504ab1d39e1611077814b2042a1a3cda5c1632300ad8.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": "5271854e133b9964ca50f9b2f7a662fbfc4451a197d4219ab3a0de0472bd15b6", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f2654c8e4368208fca03fb06da560ae6fbd57cf8054a0bcdf0769689f808d2fc", "generation_seconds_batch": 20.541158520034514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3641767501831055, "target_z": 1.0624999738266925, "ce": 5.8875885009765625, "genuineness": 4.457139015197754, "identity": 0.5237017273902893, "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/vn_STNC_reg_low/5271854e133b9964ca50f9b2f7a662fbfc4451a197d4219ab3a0de0472bd15b6.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "52d53cf6016deb08e3b9524460fd96204e091de327e58a59c90f32a269441793", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4d5cd0ed67c90053395017fa1f190433a81f0f26f3c1185f2b3b8fa79b56e656", "generation_seconds_batch": 18.043714763014577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7430145740509033, "target_z": -0.1484374116650872, "ce": 6.173524379730225, "genuineness": 4.570107460021973, "identity": 0.45486003160476685, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/52d53cf6016deb08e3b9524460fd96204e091de327e58a59c90f32a269441793.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "52e0e38f53984a57fa83a045dbbd4e66d56fea9ced6154a56d153206640e9080", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "98ef686e26aa5ef2ff8e0a0de07c03925177b02d9456e988d4e02b64274ac074", "generation_seconds_batch": 16.72890748793725, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.417551040649414, "target_z": 1.015625045803288, "ce": 5.855817794799805, "genuineness": 4.193545341491699, "identity": 0.6503332257270813, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/52e0e38f53984a57fa83a045dbbd4e66d56fea9ced6154a56d153206640e9080.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "531947c77fa650c055e3fe3b128eb31793f367af94ce531ec8894068b0e87e0b", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "91f59422c82af84f4d1b7dada30fc85d764c83457c516d6ff5338052422575ed", "generation_seconds_batch": 22.290384648018517, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.195927381515503, "target_z": 0.33203122873418767, "ce": 5.903473377227783, "genuineness": 4.557555198669434, "identity": 0.6131742596626282, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/531947c77fa650c055e3fe3b128eb31793f367af94ce531ec8894068b0e87e0b.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "5330568460152b2e09f35b9988ef2f2d814878ad7e15d38cef9ca414829008fb", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0c50dc9d4538c43626242e67d73fbf13d56208468858f058bad4db04f20a40ea", "generation_seconds_batch": 20.37721604993567, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3463852405548096, "target_z": 1.0781250196299805, "ce": 5.617537498474121, "genuineness": 4.545003414154053, "identity": 0.6948798894882202, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/5330568460152b2e09f35b9988ef2f2d814878ad7e15d38cef9ca414829008fb.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "539b6335c54794599ebeadabb013e73197f2a8688c66714d912771c710479361", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "677e9d94885aa9c19619bc181a20caae5adb1a13e00325223f6b0af81fc8bfad", "generation_seconds_batch": 22.879271298996173, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8356503248214722, "target_z": 0.6484375163583171, "ce": 6.173524379730225, "genuineness": 4.607763767242432, "identity": 0.3652637004852295, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/539b6335c54794599ebeadabb013e73197f2a8688c66714d912771c710479361.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": "53b8d303565c674f09a2751624bde070557abb580868cae1ec7db892dcd34d02", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "518684cad91c972458611e4eb18b7f47f41525687222d951a9359267a119c9da", "generation_seconds_batch": 20.41448323999066, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9246076345443726, "target_z": 0.5703124967283366, "ce": 5.903473377227783, "genuineness": 4.582659721374512, "identity": 0.6196576356887817, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het naar de lunch terug.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/53b8d303565c674f09a2751624bde070557abb580868cae1ec7db892dcd34d02.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "53bf415fb2800d51c248cab063661d3decbf6f2902965377a789fa2a73d93547", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "39a118378b5289963f5388cc2d9e6c170079f4af02a9e74afced3ce7a042db3d", "generation_seconds_batch": 16.60148960002698, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3434762954711914, "target_z": -0.6757811894742264, "ce": 6.197352409362793, "genuineness": 4.13078498840332, "identity": 0.4283776581287384, "hypothesis": "The 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/vn_STNC_reg_low/53bf415fb2800d51c248cab063661d3decbf6f2902965377a789fa2a73d93547.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "53bffecbcfe7c63a514fef49f3e98a404c35db6e5afc6f80bc8d555f8d4bc030", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "79e4449429472c47d98dc938177909c0badcf68d9adb337d188cf47a00ee01d0", "generation_seconds_batch": 17.118609820026904, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9282859563827515, "target_z": 1.4453125490749517, "ce": 5.700935363769531, "genuineness": 4.7332844734191895, "identity": 0.5293419361114502, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/53bffecbcfe7c63a514fef49f3e98a404c35db6e5afc6f80bc8d555f8d4bc030.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "6789af35b3257c6292527cd039d00d5e2944be6552d5db81fd51f9d1bfcc0c43", "generation_seconds_batch": 15.965999093023129, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.974303722381592, "target_z": -0.35156258833491283, "ce": 6.2767791748046875, "genuineness": 4.231201171875, "identity": 0.6678237915039062, "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/vn_STNC_reg_low/546203f143d2dbd908be3097fd76d41d76e93afd1487fa15c64a3d119dff5d49.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "5531a17e896a26de8592f4cc867a951a22fa09c6a8081e304d6f6f2e7c543896", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4896e1e4e125211597f53908409c9fae1e690800b96bafa945b364a695a41da0", "generation_seconds_batch": 22.61034290806856, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7644845247268677, "target_z": 0.7109374901850097, "ce": 5.784333229064941, "genuineness": 4.582659721374512, "identity": 0.6890286207199097, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/5531a17e896a26de8592f4cc867a951a22fa09c6a8081e304d6f6f2e7c543896.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "557e2cf8a1697df06ead5b8d13cfb97ec9e3a5f3820469951ed167932662efb2", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dd164d83be29cea70fd20a6f2eda4349e9a9946b2b6c8174fb843d1895e6a0cf", "generation_seconds_batch": 22.49065547704231, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3882975578308105, "target_z": 0.16308587901901606, "ce": 5.728734970092773, "genuineness": 4.570107460021973, "identity": 0.7900388240814209, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/557e2cf8a1697df06ead5b8d13cfb97ec9e3a5f3820469951ed167932662efb2.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "569caa87bb6885acef539da15f9c14496f22f2165bfa81fdf2915294a62116c4", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b9e4b150741f08eb0f107c5c4f15ab67dc3fdfc59f32bd723c4f10998b512168", "generation_seconds_batch": 17.73909819102846, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6729607582092285, "target_z": -0.08691401628775394, "ce": 6.38003396987915, "genuineness": 4.65797233581543, "identity": 0.41707223653793335, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/569caa87bb6885acef539da15f9c14496f22f2165bfa81fdf2915294a62116c4.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "56b8b9b1210b543c3aac332faa431eb694e0d719c51b7544f71ef4a615864814", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ca072efdfd919559db3f1c9d0f35825b28306ae38f60b198c90263d30cc96f2f", "generation_seconds_batch": 22.7956814939389, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3285938501358032, "target_z": 1.0937499607400387, "ce": 5.677107334136963, "genuineness": 4.469690799713135, "identity": 0.6442945599555969, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/56b8b9b1210b543c3aac332faa431eb694e0d719c51b7544f71ef4a615864814.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "571ef5973d8a547c54dc55cdf7b6ca07dade0e12e3295cb9f5324b80ec3cf4bb", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "06d5f5c07cdcc56ee977a5e5a445418f060bea4dca4c5f7af8c4582ac1aebb5d", "generation_seconds_batch": 17.0433306141058, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.049147844314575, "target_z": 0.4609374901850097, "ce": 6.157639026641846, "genuineness": 4.645420074462891, "identity": 0.7815909385681152, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt. Je fais récit.", "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/vn_STNC_reg_low/571ef5973d8a547c54dc55cdf7b6ca07dade0e12e3295cb9f5324b80ec3cf4bb.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "5732104a1eaeff52719946edc35e6766a780d50a0e8a268d10d9a1408afe7779", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9e0a7dbfbe8bab983331c47356710d075601a8b312af857ada5b236de3ac0d99", "generation_seconds_batch": 23.37151700304821, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3463852405548096, "target_z": 1.0781250196299805, "ce": 5.998785495758057, "genuineness": 4.65797233581543, "identity": 0.8329391479492188, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/5732104a1eaeff52719946edc35e6766a780d50a0e8a268d10d9a1408afe7779.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "57e7d1a8d1d118704beb700268042d7dd2070760761685c42690df300a0a3774", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4e527ab292982448b9672035acb47494f1540f6cf7052683d9b1875820480a5b", "generation_seconds_batch": 21.896026425994933, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.796389102935791, "target_z": -0.19531254907495155, "ce": 6.332377910614014, "genuineness": 4.519899368286133, "identity": 0.6265937089920044, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/57e7d1a8d1d118704beb700268042d7dd2070760761685c42690df300a0a3774.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": "58539f80841178c776da150513c6dd6f7d4de7701ee5fb584f0d328664852074", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I wrote the number on the envelope.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1d5be56e4ec29ec4e9f47215570753ddb011535b49002f54a6dac0ca2c2d91a4", "generation_seconds_batch": 21.070798308006488, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5242998600006104, "target_z": 0.9218749803700194, "ce": 5.824047088623047, "genuineness": 4.419482707977295, "identity": 0.4641534388065338, "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/vn_STNC_reg_low/58539f80841178c776da150513c6dd6f7d4de7701ee5fb584f0d328664852074.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "5855267af1483957d41a7121b354e791cb37cd324cf2e8b781b465dcb330051f", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4628314dac2d7ae221bd91d2f685daee148cbf133c7221f0b180da4a6240dd37", "generation_seconds_batch": 21.9747947860742, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.07361102104187, "target_z": 0.439453183072026, "ce": 6.0146708488464355, "genuineness": 4.632867813110352, "identity": 0.677687406539917, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/5855267af1483957d41a7121b354e791cb37cd324cf2e8b781b465dcb330051f.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": "58980dd5818889c857fcae17602a87e1315a7c74b397446ba469fafe36e6cb71", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bacb4af8c755ef5f89d5d3e8882beedb377a9b15eb80f10cba65785146a61c50", "generation_seconds_batch": 20.57339324709028, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3641767501831055, "target_z": 1.0624999738266925, "ce": 5.585766792297363, "genuineness": 4.607763767242432, "identity": 0.616893470287323, "hypothesis": "Las llaves están en la mesa de la cocina. Miren el cajón y luego, junto a la ventana, llévatelas cuando te vayas.", "wer": 0.08695652173913043, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/58980dd5818889c857fcae17602a87e1315a7c74b397446ba469fafe36e6cb71.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "59380c4e31a4ae63a026dbb4cd9551f747a8020dd143560aacd07791d283442d", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c6352757040eb5d314c5fa0a62c51f7fe95c347c25a5647e16cc2b17bbda5545", "generation_seconds_batch": 21.786103532998823, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5509870052337646, "target_z": 0.8984375163583171, "ce": 6.300607204437256, "genuineness": 4.419482707977295, "identity": 0.23533904552459717, "hypothesis": "The 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/vn_STNC_reg_low/59380c4e31a4ae63a026dbb4cd9551f747a8020dd143560aacd07791d283442d.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "5958d83e8f4df348e81164f4165b8524098d86a15113b9bab7e16525b5048ba0", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "07d5c5b3835176cfe2a0bf488cec77622d31ca74fa858540bbb0dde3b4b30184", "generation_seconds_batch": 22.749273675959557, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7644845247268677, "target_z": 0.7109374901850097, "ce": 5.736677646636963, "genuineness": 4.105680465698242, "identity": 0.48615333437919617, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/5958d83e8f4df348e81164f4165b8524098d86a15113b9bab7e16525b5048ba0.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "59e49468df9960f403757ef9dc8590bb2431060108c9bb119f27cbb056e73467", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4670e6000836d970eba10480316565b57d56a94fbc5f10ee2db60ba35f83fbb4", "generation_seconds_batch": 16.030564768007025, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3960812091827393, "target_z": 0.15625003925996125, "ce": 6.0146708488464355, "genuineness": 4.68307638168335, "identity": 0.836263120174408, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/59e49468df9960f403757ef9dc8590bb2431060108c9bb119f27cbb056e73467.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "5ad25efc1259a4b7942c2b63bcb9a1534198201d5904e8c772590ababd0dd25e", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c945c11533c860d9c1225e9ead8431b62dc55df571ea2b43bb1b1ccca0e02961", "generation_seconds_batch": 16.605105969007127, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5420912504196167, "target_z": 0.9062500392599613, "ce": 5.6334228515625, "genuineness": 4.65797233581543, "identity": 0.6628413796424866, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/5ad25efc1259a4b7942c2b63bcb9a1534198201d5904e8c772590ababd0dd25e.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "5afeca86380c44f8c642856970657d3a303b528530ac6de082e7ea3603f4cc50", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "28733f52b2846f48dc50fee11d162a7b446d5df8d5f3cd1466c6a956248b8340", "generation_seconds_batch": 22.038143103942275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.079513430595398, "target_z": 1.3124999738266925, "ce": 5.649308204650879, "genuineness": 4.883909225463867, "identity": 0.7841888666152954, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/5afeca86380c44f8c642856970657d3a303b528530ac6de082e7ea3603f4cc50.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "5b02a0da660c99fefa2c73e55ebddd9b94bc8d9be29456623691583d70bd8635", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "7f2b0e13e6a9538ecbfe9d65b0b25a8317947d84f78514052ef224b711670278", "generation_seconds_batch": 23.269974588998593, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.387185573577881, "target_z": 0.1640624574683753, "ce": 6.078212261199951, "genuineness": 3.8922953605651855, "identity": 0.31126970052719116, "hypothesis": "The 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/vn_STNC_reg_low/5b02a0da660c99fefa2c73e55ebddd9b94bc8d9be29456623691583d70bd8635.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "5b2c2dc9297043a2083109e94a195a7b60107facd1c4ec1d6b02b2072579d4b2", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bc6427d2047a1746e915dacbe5452b0fff1637e36743bf22d5f1c0992c7bb923", "generation_seconds_batch": 22.142331150942482, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7733802795410156, "target_z": 0.7031249672833656, "ce": 6.3641486167907715, "genuineness": 4.720732688903809, "identity": 0.5499555468559265, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/5b2c2dc9297043a2083109e94a195a7b60107facd1c4ec1d6b02b2072579d4b2.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "5b812c935694bf813a37c094d8543e96a88a09bd0eb5ff76c2efeff4215037db", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "fe0ac401e1382045b2ab19d917d774bc0dfd2baa0d2c6986ea8cf88dd82657f2", "generation_seconds_batch": 16.60148960002698, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3293633460998535, "target_z": 0.21484369928921673, "ce": 6.109982967376709, "genuineness": 4.13078498840332, "identity": 0.4311922788619995, "hypothesis": "The 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/vn_STNC_reg_low/5b812c935694bf813a37c094d8543e96a88a09bd0eb5ff76c2efeff4215037db.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "5be4e302ef0764f07fab9c4a2659c734254f05914cf1f4b7d9beb2628e22278e", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "058317bcdb1830f3c64c3aae800e25b0183a6baa209929bba9ce08ddbde3a58d", "generation_seconds_batch": 23.837759347981773, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5598827600479126, "target_z": 0.8906249934566731, "ce": 6.038498878479004, "genuineness": 4.7458367347717285, "identity": 0.6252021789550781, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/5be4e302ef0764f07fab9c4a2659c734254f05914cf1f4b7d9beb2628e22278e.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "5bec52cbf563c8d2b7e6d8a20ffd44875e5c7f81471c64941daaba3fd70a21ba", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2b3bdda46e793c354fbe34a2ddf8f88207f804b3b9aad2d53a97b2fd14087bd8", "generation_seconds_batch": 22.854234994971193, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2307407855987549, "target_z": 1.1796875032716634, "ce": 5.784333229064941, "genuineness": 4.507347106933594, "identity": 0.36853039264678955, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/5bec52cbf563c8d2b7e6d8a20ffd44875e5c7f81471c64941daaba3fd70a21ba.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "5c3d75d8cfd56d7bea99eceef953df88aa8f59231d1921f6996e0dce386b5e30", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "775f57fe12063e426ca3f16e0d83ad683c01f812cb6d37e5688b5fe413ebc39f", "generation_seconds_batch": 16.889224926009774, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.035804271697998, "target_z": 0.4726562221908608, "ce": 6.1894097328186035, "genuineness": 4.281409740447998, "identity": 0.35756126046180725, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/5c3d75d8cfd56d7bea99eceef953df88aa8f59231d1921f6996e0dce386b5e30.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "5c8e4b9ba41f8257778ededd974dbc3fcbeefd53def10a3a4171ec70a74b7ada", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e500d13a7cf7490c6678aaf41c0f00ebe541bc59b8368c9e8ad55b12b445aa28", "generation_seconds_batch": 23.733957962016575, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4976125955581665, "target_z": 0.9453125490749515, "ce": 5.879645347595215, "genuineness": 4.432034492492676, "identity": 0.5479627847671509, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/5c8e4b9ba41f8257778ededd974dbc3fcbeefd53def10a3a4171ec70a74b7ada.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "5cd125e01da73c7a5056ed4e4ca0f085838e1977d509132017e6bac1b0b56f97", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "059ea2d6c759f9267e0bf474aa1304469ac72c6d8e246cdb5fe4ac0909629c56", "generation_seconds_batch": 21.61453690403141, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.608809232711792, "target_z": 0.8476562745374758, "ce": 5.879645347595215, "genuineness": 4.6705241203308105, "identity": 0.5203288197517395, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/5cd125e01da73c7a5056ed4e4ca0f085838e1977d509132017e6bac1b0b56f97.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "5ce432c11e5f080348ed891a0bb572a8572d34408dd3ddac636b76858fe660ad", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b370c8626915ab4d6f591eb4510ba32ff42eb3ee3363b6d491ff7b9cc44c63a3", "generation_seconds_batch": 23.34666066803038, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3285938501358032, "target_z": 1.0937499607400387, "ce": 5.613565921783447, "genuineness": 4.507347106933594, "identity": 0.6631503701210022, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/5ce432c11e5f080348ed891a0bb572a8572d34408dd3ddac636b76858fe660ad.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "5d683f4cd7d4dc71c9876cb982ccbbaa6c882dce0625890ebd8ae7906f26b39e", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "52f4f65098a746eceff3ec8906c014495834aed7e7ad6bd79fff98a0aec4b7f2", "generation_seconds_batch": 22.10517268104013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.018012762069702, "target_z": 0.4882812679941489, "ce": 5.736677646636963, "genuineness": 4.507347106933594, "identity": 0.8262172937393188, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/5d683f4cd7d4dc71c9876cb982ccbbaa6c882dce0625890ebd8ae7906f26b39e.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "5d7d9142c019e798c6ffca8a3815973bfda9977be5330897fc6764d461223486", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0acba414bac42a73d30bf70b4a2ca86267bd40a78bcd2c5176df1addf1609401", "generation_seconds_batch": 16.871221506036818, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.092172145843506, "target_z": -0.45507822887531413, "ce": 6.340320587158203, "genuineness": 4.2374773025512695, "identity": 0.15989325940608978, "hypothesis": "The 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/vn_STNC_reg_low/5d7d9142c019e798c6ffca8a3815973bfda9977be5330897fc6764d461223486.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "5d8709d767506fc0c79f2f31190adaf8259ffa7505e0270676c5242bec7cb5ad", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1c12c4ebe5a9ef318ff845afe223ec9ac874dd3b6e2f1b2815a862b9a1249fa8", "generation_seconds_batch": 22.393496081000194, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.49239540100097656, "target_z": 1.8281249149367507, "ce": 5.855817794799805, "genuineness": 4.6705241203308105, "identity": 0.29707908630371094, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/5d8709d767506fc0c79f2f31190adaf8259ffa7505e0270676c5242bec7cb5ad.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "5e00f8517dc4ae9470cd1bcae6362a0561f64603d4bbf3b1043339d306aae89c", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b39dbf3aa5743cf5dd8e15ff534dd8e65ca9eba49e9422bf81e11227f669ac73", "generation_seconds_batch": 17.107407317962497, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8178589344024658, "target_z": 0.6640624574683753, "ce": 5.824047088623047, "genuineness": 4.6956281661987305, "identity": 0.7472535371780396, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/5e00f8517dc4ae9470cd1bcae6362a0561f64603d4bbf3b1043339d306aae89c.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "c9fd66331f5e642e037bafbf3080c958e5704503476ef12bbcef8b4adc8bab90", "generation_seconds_batch": 16.457021304988302, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.115865707397461, "target_z": 0.4023438301557542, "ce": 6.09409761428833, "genuineness": 4.582659721374512, "identity": 0.8248690962791443, "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/vn_STNC_reg_low/5e0af124ad7421917e897f2876d49b25e213a1bc87e01fe43f09b9c255c2a32f.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "5e489a1bddc7ce914d764ab34ea21087560cf1c0a15a8aa9200329aa0ba47f25", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "84a8442bbfa41c01d40d17083784409e1cd93f65bf54d85466494b36c3c89c9d", "generation_seconds_batch": 22.49065547704231, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0869545936584473, "target_z": 0.4277344510661749, "ce": 5.982900142669678, "genuineness": 4.2751336097717285, "identity": 0.499822199344635, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/5e489a1bddc7ce914d764ab34ea21087560cf1c0a15a8aa9200329aa0ba47f25.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "5f6d64eb214db2e0978ffdfebf8c5e941a7960fd89eece8e40d515ce21db4ac2", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2c99109ad4ae430513d99c4fb2210ed33c6940eca58aecaa31d7106c01e0d89e", "generation_seconds_batch": 22.81780348799657, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2181665897369385, "target_z": 0.3125000785199225, "ce": 6.229123115539551, "genuineness": 4.6956281661987305, "identity": 0.47467976808547974, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/5f6d64eb214db2e0978ffdfebf8c5e941a7960fd89eece8e40d515ce21db4ac2.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "601dfb59d8a91ae1f17656dde51690d6c18d9597fcd94cfc31005ab7b9af34fd", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cd5e6497c2b9da9f7829efdd79ce30c04af2ec67b7816a3eb915100626a3fc9e", "generation_seconds_batch": 23.84609296103008, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4086554050445557, "target_z": 1.023437464011702, "ce": 5.911416053771973, "genuineness": 4.8588056564331055, "identity": 0.615668535232544, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt ?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/601dfb59d8a91ae1f17656dde51690d6c18d9597fcd94cfc31005ab7b9af34fd.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": "6054c4c3578641a7b9820e5d7a33d38ac2324ff416938ace5cd422c17fe44875", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bbd2f765df466ae37cdcc4f5689983b1b4f756f41dbb0620a96bed0a8607cf0f", "generation_seconds_batch": 16.662871649954468, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3819681406021118, "target_z": 1.0468750327166343, "ce": 5.824047088623047, "genuineness": 4.582659721374512, "identity": 0.359427273273468, "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/vn_STNC_reg_low/6054c4c3578641a7b9820e5d7a33d38ac2324ff416938ace5cd422c17fe44875.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "60c3b515eaab5ce4cc1add233c2a8a5fba6f1eb6f665afcb2e659a46bda69f8b", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6b2b0a2a0707a49b087ac9d2d7f12c2c416ae2851f06edd88172b0642f809229", "generation_seconds_batch": 22.527320639928803, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.473362922668457, "target_z": 0.08837888396179283, "ce": 6.062326908111572, "genuineness": 4.457139015197754, "identity": 0.5293534398078918, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/60c3b515eaab5ce4cc1add233c2a8a5fba6f1eb6f665afcb2e659a46bda69f8b.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "618cb2541ebccfc73628696533bf4d1b4a492f96008d4cc83c49792b6764f694", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f87e74f7f230b9a1d0e45f85519922536d71bbb133d19485bd6f3a1ed06b2e1d", "generation_seconds_batch": 16.873286899994127, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4531340599060059, "target_z": 0.9843749541967118, "ce": 5.784333229064941, "genuineness": 4.595211505889893, "identity": 0.6084959506988525, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/618cb2541ebccfc73628696533bf4d1b4a492f96008d4cc83c49792b6764f694.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "61902bf218559bde81fd89736ec8f7fd7f60bc0b8519a6a73047234bbc59a929", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "697252b7d9e355f1fa6d501bcc8c43c9b61b718d644080d32e34d6a7846c8917", "generation_seconds_batch": 17.450747145107016, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3149077892303467, "target_z": 0.22753900974442706, "ce": 6.173524379730225, "genuineness": 4.7081804275512695, "identity": 0.6537191867828369, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/61902bf218559bde81fd89736ec8f7fd7f60bc0b8519a6a73047234bbc59a929.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "62733c82558d340d69093b9518624a3f0e51e0844bef938df9fc323913153047", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ae240831bf396e6a04d052f5b6f2bbd303b5e96d4761d15af714d32806202f92", "generation_seconds_batch": 17.433165887952782, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3716180324554443, "target_z": 0.1777343463729449, "ce": 6.0305562019348145, "genuineness": 4.356721878051758, "identity": 0.6635797023773193, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/62733c82558d340d69093b9518624a3f0e51e0844bef938df9fc323913153047.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "6292ffc8d7d89705eab8cdba3a0f1d3ca00f00e88a7fcc906b2e1f727f0486af", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1f307f03f92e10ae6d16df0f913377a4a807b301b7b08fab7537a6549738b66b", "generation_seconds_batch": 23.382360587012954, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.925377130508423, "target_z": -0.30859376472248545, "ce": 5.967014789581299, "genuineness": 4.055472373962402, "identity": 0.3440927267074585, "hypothesis": "The 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/vn_STNC_reg_low/6292ffc8d7d89705eab8cdba3a0f1d3ca00f00e88a7fcc906b2e1f727f0486af.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "62bb85188c8bdca5a0f7a2b92ce0cdb792093c2269ce3f5df844e41b5f2096a8", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "14f52866c8d38714eb11c3fcfd5f49b0f88ffa7a30f43998f85c7a22e816f98c", "generation_seconds_batch": 23.609800353995524, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6443922519683838, "target_z": 0.8164061829308995, "ce": 5.7247633934021, "genuineness": 4.3127899169921875, "identity": 0.4263085722923279, "hypothesis": "The door is still open. I put your coat on the chair and tell me when you are ready.", "wer": 0.05555555555555555, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/62bb85188c8bdca5a0f7a2b92ce0cdb792093c2269ce3f5df844e41b5f2096a8.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": "6306fb2449164d29b0110bd2e017bb792c2c942dfb5bc3ba4d192615da1088d2", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fd9d7cddcc7c22dc3aee9b712c06b0ce7e34d90ea588379e18f79ca0ce88f925", "generation_seconds_batch": 15.84039844200015, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5865700244903564, "target_z": 0.8671874247517409, "ce": 5.546053409576416, "genuineness": 4.632867813110352, "identity": 0.734664261341095, "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/vn_STNC_reg_low/6306fb2449164d29b0110bd2e017bb792c2c942dfb5bc3ba4d192615da1088d2.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "635e2dec76b8f0f9e7319dc110e2c43f8aa85e9d429762b920def3eae4b542d6", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "000f89a29726851e64c9326a9affa5a478b1d19ab83e46bebb058b2038ab713b", "generation_seconds_batch": 16.845196036971174, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.20927095413208, "target_z": 0.32031249672833656, "ce": 6.062326908111572, "genuineness": 4.582659721374512, "identity": 0.7157415151596069, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/635e2dec76b8f0f9e7319dc110e2c43f8aa85e9d429762b920def3eae4b542d6.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": "6369526e4933b42b0076dc5b5b142415540f5a6c891c44689949b0fad958f9d4", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I wrote the number on the envelope.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9931ccdd564fb28588416fd70ea9dcac86e08ef7e56eed78b0c8d800711e35d6", "generation_seconds_batch": 17.52171370398719, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1492247581481934, "target_z": 0.3730468954478965, "ce": 6.23706579208374, "genuineness": 4.519899368286133, "identity": 0.44977232813835144, "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/vn_STNC_reg_low/6369526e4933b42b0076dc5b5b142415540f5a6c891c44689949b0fad958f9d4.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "637ef3138b0105c4ff66546f82aa2cbd996642ad945a9c79281425891f7d4dd6", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c98d354e9235ffed9ff50898b33a3b219f65e77bbb0f5f62350c88558a206db9", "generation_seconds_batch": 26.79255271994043, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4620296955108643, "target_z": 0.9765625359882978, "ce": 5.414999485015869, "genuineness": 4.770940780639648, "identity": 0.6131085157394409, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/637ef3138b0105c4ff66546f82aa2cbd996642ad945a9c79281425891f7d4dd6.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "638e19f3394cd764c6b2cfb4d6db63604e8267be8ef2796c6def498dff6e8cca", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "948fdb1822c48cd24f5a2c9c9496056894dc463d52bc46087ccf94aef82b95c8", "generation_seconds_batch": 17.497616717009805, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1566660404205322, "target_z": -0.511718732005851, "ce": 6.403861999511719, "genuineness": 4.369274139404297, "identity": 0.6815582513809204, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/638e19f3394cd764c6b2cfb4d6db63604e8267be8ef2796c6def498dff6e8cca.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": "653c6d7cc9145db4cdf19500edfa2646d4cd9900b944b6977ee9cbc0fe3f4f18", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "26755971a2e03b5e8431038d88b161545750a286d4dc459d6797bbb6f1b6dba4", "generation_seconds_batch": 17.83537497301586, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7377972602844238, "target_z": 0.7343750588899419, "ce": 6.038498878479004, "genuineness": 4.632867813110352, "identity": 0.8340876698493958, "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/vn_STNC_reg_low/653c6d7cc9145db4cdf19500edfa2646d4cd9900b944b6977ee9cbc0fe3f4f18.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "65aeb79ec6963c8bc2eab7fe255f7b26108dd40ff18aeb40bc5b3922ecd96b8d", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "63dcb29b0bd24813509d6e4a5b76b075fbf2cb279a572d771a517532e99b8495", "generation_seconds_batch": 20.960406412021257, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.782275915145874, "target_z": 0.6953125490749515, "ce": 5.919358730316162, "genuineness": 4.557555198669434, "identity": 0.6583492755889893, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/65aeb79ec6963c8bc2eab7fe255f7b26108dd40ff18aeb40bc5b3922ecd96b8d.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "65c32fbf5a3ddaa2500ed0e09bb897ee5a0b8a970949292ce73cccf6add0a17b", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "caf6aa1e5b6f148dcdf5cfa951ee38eadf5197ac545a3bd20858df2d27041d09", "generation_seconds_batch": 17.5902086350834, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.34493088722229, "target_z": 0.20117181038464713, "ce": 5.871703147888184, "genuineness": 4.118232727050781, "identity": 0.518997073173523, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/65c32fbf5a3ddaa2500ed0e09bb897ee5a0b8a970949292ce73cccf6add0a17b.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "65cd0b4797f1c0b10fc7b0abdbbcffe853e2f0114f0c9d465d7da7c45f02039f", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "958bd43d689f914a2bcc7e201c98c5c1b4a02fb25130a69cf6c82021e2945eb4", "generation_seconds_batch": 22.692677541985177, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8267545700073242, "target_z": 0.6562500392599613, "ce": 6.125868320465088, "genuineness": 4.68307638168335, "identity": 0.753595232963562, "hypothesis": "La porte était encore ouverte. J'ai posé ton manteau sur la j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.3888888888888889, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/65cd0b4797f1c0b10fc7b0abdbbcffe853e2f0114f0c9d465d7da7c45f02039f.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "65db3e57fd80a4300f0ed98050899165929026c732d1bcd6378c97a62de944e4", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ebfdf277cfaef478893aa4789ff89fc3fd10edb39691c6563dac51f239197de5", "generation_seconds_batch": 16.47657114895992, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.008347511291504, "target_z": 1.375000052346615, "ce": 5.903473377227783, "genuineness": 4.959221839904785, "identity": 0.7623158097267151, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/65db3e57fd80a4300f0ed98050899165929026c732d1bcd6378c97a62de944e4.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "65ef4b55f90dcc02d91e78b0587ea765db406a8c3c76d707f8dcd89b440974d1", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b4fad91dd455592fa8c827a7c8c75aad92e5feb52b03ba398c5cd22c4546947f", "generation_seconds_batch": 16.308418160071597, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.079513430595398, "target_z": 1.3124999738266925, "ce": 5.355429172515869, "genuineness": 4.796044826507568, "identity": 0.7028752565383911, "hypothesis": "La porte était encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/65ef4b55f90dcc02d91e78b0587ea765db406a8c3c76d707f8dcd89b440974d1.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "670f4f2149ed7666706b5b257c58f31b4b060f518ed2193253d5a318251585ed", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "12e3cf77a8dccfc089ed03a301e8fc40e5a677a443696bdf07f121f76ef46a80", "generation_seconds_batch": 17.175231553963386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9824298620224, "target_z": 0.5195312549074952, "ce": 5.974957466125488, "genuineness": 4.65797233581543, "identity": 0.6181722283363342, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/670f4f2149ed7666706b5b257c58f31b4b060f518ed2193253d5a318251585ed.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "67135479cc0cf63eea75f85cd4255fb120d5ce9d081038007df4cd1f9a43d6f1", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e8697c2a5be68109857fe38f501ec811f85752c4fae25202238f954a2db7b3b2", "generation_seconds_batch": 15.981148398015648, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3196980953216553, "target_z": 1.1015624836416829, "ce": 5.847874641418457, "genuineness": 4.595211505889893, "identity": 0.2745332717895508, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/67135479cc0cf63eea75f85cd4255fb120d5ce9d081038007df4cd1f9a43d6f1.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "6744fda66e7b0353bca26a8297bb59633b42424a3088310df896e2d02200cd1e", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c28e39d6d043f6ccb339c7c20c5ab77acc6bc81d4f7cfe8d999fa1f8d831a5a0", "generation_seconds_batch": 22.877725863014348, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.648840069770813, "target_z": 0.8124999738266925, "ce": 5.677107334136963, "genuineness": 4.469690799713135, "identity": 0.6135361790657043, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/6744fda66e7b0353bca26a8297bb59633b42424a3088310df896e2d02200cd1e.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "675d6c7d9d7f02e84c8aebb02e043150761b9c7bb079385d9b55ff19338d2141", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d78220fa523087b818b9ad50c6886f1692c1c1afb056cc30dd4733f1014981d3", "generation_seconds_batch": 22.056031152023934, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9112640619277954, "target_z": 0.5820312287341877, "ce": 6.157639026641846, "genuineness": 4.469690799713135, "identity": 0.3263091742992401, "hypothesis": "The 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/vn_STNC_reg_low/675d6c7d9d7f02e84c8aebb02e043150761b9c7bb079385d9b55ff19338d2141.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "681470089e163c8f99a49de2421074b908a822b5e06da67af2ea1bebac144bc6", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "51faab1835858bd99d111bf513e985568c0fc85c2c121184a907a5a1a3f5e2be", "generation_seconds_batch": 22.9115524640074, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.691864252090454, "target_z": -0.10351564054040133, "ce": 6.300607204437256, "genuineness": 4.425758361816406, "identity": 0.538943350315094, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/681470089e163c8f99a49de2421074b908a822b5e06da67af2ea1bebac144bc6.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "684320feacf9b79d2033a552250336943ebd799b52e7115aad95c07b129769ad", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3296e70debcb43273d9222867b1f54dac2837b06bab49673d0293fc1ac11d9fe", "generation_seconds_batch": 22.017901215003803, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1417834758758545, "target_z": 1.2578125229016441, "ce": 5.80021858215332, "genuineness": 4.770940780639648, "identity": 0.547702431678772, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/684320feacf9b79d2033a552250336943ebd799b52e7115aad95c07b129769ad.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "68b1c810550a9cf035c7d9dd8012d9a0c68d1f9ae4a71e00870623ff9ea07841", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "86ef82f7bed7a8a8a134d26b65e8fa1a6199db02540772e348429ca9d3e3b56e", "generation_seconds_batch": 22.02541480807122, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3285938501358032, "target_z": 1.0937499607400387, "ce": 5.736677646636963, "genuineness": 4.68307638168335, "identity": 0.7907407283782959, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/68b1c810550a9cf035c7d9dd8012d9a0c68d1f9ae4a71e00870623ff9ea07841.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "68f3cb55d01cfee6e4a0f8c516d86c5e0a220d0d1a239d043ef7952c854741d7", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f6bdee6b002d993751f2974eb745b318e519fbcbd585d57bb40550c937382694", "generation_seconds_batch": 15.779992735944688, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4683589935302734, "target_z": 0.09277348698390948, "ce": 6.2767791748046875, "genuineness": 4.406930446624756, "identity": 0.37811484932899475, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/68f3cb55d01cfee6e4a0f8c516d86c5e0a220d0d1a239d043ef7952c854741d7.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "694549a03325dd4c846810ce12247d91138bcf8324d9d3a8b848a234017f18c0", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "61bbb5efcca7920c66f6714e16cbe282743649393644c95f220537124a2a2a4c", "generation_seconds_batch": 23.836005758028477, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7948501110076904, "target_z": 1.5624999738266925, "ce": 5.363371849060059, "genuineness": 4.808597087860107, "identity": 0.6835691332817078, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/694549a03325dd4c846810ce12247d91138bcf8324d9d3a8b848a234017f18c0.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "697f47e40ecee322a0e1fe2da2acbd523c3fde2ec992f7b97702b963cb1ea403", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c9fc5c4895d6a79f0b6bbad61ef5105c680e1514e917b9231e6a6ef8b3839347", "generation_seconds_batch": 16.39742056292016, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3894095420837402, "target_z": 0.1621093005696568, "ce": 6.062326908111572, "genuineness": 4.758388519287109, "identity": 0.7803144454956055, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/697f47e40ecee322a0e1fe2da2acbd523c3fde2ec992f7b97702b963cb1ea403.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "69a543df36309abeaa3a3550521ad5d84a422f1dbf2c68f0edbddea40c4e6795", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "5b778fb95f48615deae36b25a6ea7c1e4ae233306d2c04a28b90708d36bed0db", "generation_seconds_batch": 23.169558982015587, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4883744716644287, "target_z": 0.07519528428190285, "ce": 6.316492557525635, "genuineness": 4.432034492492676, "identity": 0.21367910504341125, "hypothesis": "The 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/vn_STNC_reg_low/69a543df36309abeaa3a3550521ad5d84a422f1dbf2c68f0edbddea40c4e6795.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "69c4694df560acd741d01971260439f01da7547570fa7f82845ad4ed2a565772", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b256c27e84a1c352dbb088c350336ee802a62f8b6819d1a360595c82ac1d1a15", "generation_seconds_batch": 23.271698817028664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8126415014266968, "target_z": 1.5468750327166343, "ce": 5.824047088623047, "genuineness": 4.7081804275512695, "identity": 0.39014148712158203, "hypothesis": "I found the little box. It was beside the old window. You can look inside together.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/69c4694df560acd741d01971260439f01da7547570fa7f82845ad4ed2a565772.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "6a0d2e4f1b51b36d739526e5657288b441637a341ef23ce75be937dea164f7cf", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7054be382fd1394ceb9a0600472efd12b7f2bdcf977983a276521fde6e1d9117", "generation_seconds_batch": 22.680006399052218, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0172432661056519, "target_z": 1.367187529444971, "ce": 6.046441555023193, "genuineness": 4.632867813110352, "identity": 0.6611826419830322, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/6a0d2e4f1b51b36d739526e5657288b441637a341ef23ce75be937dea164f7cf.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "6a55433664b5fa9a26836f3fbcc8d8da7610431a85be8fd6a990bdbefef3624c", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "74739ef9063d35d702a241fa93ac0eb51ae44628a3f5a258b2c29503b08706f7", "generation_seconds_batch": 16.752372315968387, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7155580520629883, "target_z": 0.753906209104207, "ce": 5.95112943649292, "genuineness": 4.507347106933594, "identity": 0.8037077784538269, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/6a55433664b5fa9a26836f3fbcc8d8da7610431a85be8fd6a990bdbefef3624c.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "4b58a8e63029e8463848ce651827578db76836a72b9c0bab5d38026ef2128cb3", "generation_seconds_batch": 16.82580221700482, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.298997640609741, "target_z": -0.6367186796592361, "ce": 6.221180438995361, "genuineness": 4.281409740447998, "identity": 0.4093274772167206, "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/vn_STNC_reg_low/6a61bc20d43b3e97640d8dd45682f7055c5dbaa5aa987cdb55f8a7d7a432a2c2.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "6b5fa592a2ad9db7f4b6112dccb0c1c6eee1466812c7e5538c40be5b7b320ecd", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d51655e1fe67ef73cb885b175eafbd66a28765d5d479d04f5ec981ad1f24c431", "generation_seconds_batch": 17.007239979924634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2226145267486572, "target_z": 0.30859376472248545, "ce": 5.998785495758057, "genuineness": 4.444586753845215, "identity": 0.5336846709251404, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/6b5fa592a2ad9db7f4b6112dccb0c1c6eee1466812c7e5538c40be5b7b320ecd.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "4e020fa24be632253dbb81908dfbcd5387d40b8979056a68e7f8076fce348210", "generation_seconds_batch": 17.193959300988354, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6251463890075684, "target_z": -0.04492198051114585, "ce": 6.173524379730225, "genuineness": 4.607763767242432, "identity": 0.6453262567520142, "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/vn_STNC_reg_low/6b973c750e3616c70f28663c29496a0c74bea4a87dff4bad65cadc5394b99131.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": "6b9a39eb05e72aec8396c2dd800e7ffc7eea618521d0c2c7d7927ee22b996a49", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Escribí el número en el sobre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "814753f5b4258ef65753be233daaef326e048b14ff78644bee2bc51a3cc957aa", "generation_seconds_batch": 20.4698285079794, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8445460796356201, "target_z": 0.6406249934566731, "ce": 5.816103935241699, "genuineness": 4.8588056564331055, "identity": 0.674795389175415, "hypothesis": "Escribí el número en en el sobre, léelo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/6b9a39eb05e72aec8396c2dd800e7ffc7eea618521d0c2c7d7927ee22b996a49.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "6be50c5134ec74aa030f4aa6cce6f5a3b0a3108244b9d6338341523e625cbae2", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0a2726eaae8b61e0089a35b6edf97a59b9e9991632554d61e37c7cce9686695c", "generation_seconds_batch": 23.694935475010425, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.848224401473999, "target_z": 1.515625045803288, "ce": 5.665193557739258, "genuineness": 4.632867813110352, "identity": 0.45342397689819336, "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/vn_STNC_reg_low/6be50c5134ec74aa030f4aa6cce6f5a3b0a3108244b9d6338341523e625cbae2.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "6c58115a8c0206d940daab2576c7d629c915afe99eda8c37a94c0a55ab7adde6", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fea6209e2b955abfb3d6d9f2fadba244910011e7e01b4ca0a1ff6a164be0f0cf", "generation_seconds_batch": 16.296801679069176, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.775261640548706, "target_z": -0.17675776792358566, "ce": 6.38003396987915, "genuineness": 4.6705241203308105, "identity": 0.4878910779953003, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/6c58115a8c0206d940daab2576c7d629c915afe99eda8c37a94c0a55ab7adde6.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "6c5a16bd332646c3bd8085f0e991ccd00090dd239b4c151e59d4cbfc0fcb42a4", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d563c3e24c5187ca1d6ed831d42a639d3b626e38d875e7e87c83e3097d60331d", "generation_seconds_batch": 23.99642169906292, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.643622636795044, "target_z": 1.6953125490749517, "ce": 5.228346347808838, "genuineness": 4.7081804275512695, "identity": 0.6568192839622498, "hypothesis": "La puerta sigue abierta. He dejado tu abuelo 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/vn_STNC_reg_low/6c5a16bd332646c3bd8085f0e991ccd00090dd239b4c151e59d4cbfc0fcb42a4.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "6c8ca28b5c22ce7a14c5c20b7536ede55f14988d5eb3f8c4426bcc7c8efee330", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "13d516d836fcddebd8790ed8ae1989de45a601dc8539ce2a3c64547edfc8ba35", "generation_seconds_batch": 22.535716219921596, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0973048210144043, "target_z": 1.2968750327166343, "ce": 5.649308204650879, "genuineness": 4.6203155517578125, "identity": 0.5554084777832031, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/6c8ca28b5c22ce7a14c5c20b7536ede55f14988d5eb3f8c4426bcc7c8efee330.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "6d4069052e81f9caab99651f3448e34e28a7cf848d4dd319e7258602c3a3478d", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ab898d8bf185d4ce3bb225533137189153978bcd16de915c58be1e30cce331aa", "generation_seconds_batch": 18.62882692704443, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.820852279663086, "target_z": -0.21679685618793523, "ce": 6.229123115539551, "genuineness": 4.557555198669434, "identity": 0.666345477104187, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/6d4069052e81f9caab99651f3448e34e28a7cf848d4dd319e7258602c3a3478d.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "6d55f4299269bf3bfcc3c3fa0d6fe2579296fb392197630fd0c9b8713ba9688a", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "31ac22b80956510d285164516bc378bc2274ad4a8dab02e3ff402b01c99edbc1", "generation_seconds_batch": 17.21068907994777, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.236727714538574, "target_z": -0.5820313334274176, "ce": 6.229123115539551, "genuineness": 4.18099308013916, "identity": 0.3399116098880768, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/6d55f4299269bf3bfcc3c3fa0d6fe2579296fb392197630fd0c9b8713ba9688a.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "6dba0ae46cb5b219a28b77756d380d98bd2d3ffcdffa1b439fc054cfabf763a7", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6f3b931dd195807ba1425964edc2c677a12513ff5ac8cf92d642c4e2f1dce914", "generation_seconds_batch": 16.715432905009948, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7663660049438477, "target_z": -0.1689453497151716, "ce": 6.109982967376709, "genuineness": 4.519899368286133, "identity": 0.3539327383041382, "hypothesis": "ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/6dba0ae46cb5b219a28b77756d380d98bd2d3ffcdffa1b439fc054cfabf763a7.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "6de8966bec375da4ef16cd0d979a495c5fdb2cf9c88eeb7dd948a6e5ba558dc3", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "67e95e1bd97b031c644aeace6519162f0bbea47e563ed6a65d91eee58cec015e", "generation_seconds_batch": 21.9747947860742, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3960812091827393, "target_z": 0.15625003925996125, "ce": 6.078212261199951, "genuineness": 4.632867813110352, "identity": 0.6556751132011414, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/6de8966bec375da4ef16cd0d979a495c5fdb2cf9c88eeb7dd948a6e5ba558dc3.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "6e39aeba1b1dcdc04549251affb9ae86253e395cea1ed698ec264b811e274a72", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3ba41a14e67a27b0d25ed90a6c22c98430a646b089acdf99c17088e6910d4e20", "generation_seconds_batch": 16.49756517598871, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.189255475997925, "target_z": 0.3378906994303432, "ce": 6.165581703186035, "genuineness": 4.243753433227539, "identity": 0.40218889713287354, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/6e39aeba1b1dcdc04549251affb9ae86253e395cea1ed698ec264b811e274a72.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "6f0b57227d6df6700f24995feb73f364755f857200fbdaff4cb1772906e6e480", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d7ed7039cebf86d0ff8da2563d5c27250333d37be6954021300ed5506e21187d", "generation_seconds_batch": 17.184933942044154, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.820852279663086, "target_z": -0.21679685618793523, "ce": 6.308549880981445, "genuineness": 4.356721878051758, "identity": 0.6898667216300964, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/6f0b57227d6df6700f24995feb73f364755f857200fbdaff4cb1772906e6e480.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "6f0e24a2fa331e3abd15f6d30dd8bb9c9e7152227bab0f4bf705348f6c435fb0", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e7df29c1cef3fd7a0954ae4a2bfc0c90aa9499d9f6a6350248e37951e6a210fa", "generation_seconds_batch": 23.868397870915942, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2396365404129028, "target_z": 1.1718749803700195, "ce": 5.593709468841553, "genuineness": 4.8588056564331055, "identity": 0.594734787940979, "hypothesis": "He encontrado la cajita, estaba junto a la ventana vieja, ¿podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/6f0e24a2fa331e3abd15f6d30dd8bb9c9e7152227bab0f4bf705348f6c435fb0.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": "6f171df18b14eb7ebf82fa07f5a2844a4506843341ae1dd7e694f1a54867f62a", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The road turns left behind the bridge.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5f688ae6b3162f7fcd4c4ed97582518442c315692daccdce960aa9bf16789aee", "generation_seconds_batch": 22.58502525393851, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6844229698181152, "target_z": 0.7812499869133462, "ce": 5.7684478759765625, "genuineness": 4.3881025314331055, "identity": 0.5820952653884888, "hypothesis": "The road turns left behind the bridge. You keep going until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.041666666666666664, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/6f171df18b14eb7ebf82fa07f5a2844a4506843341ae1dd7e694f1a54867f62a.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "6f1c866210bbdda9e4559a6d5ce8c4083c35d72fb6d88dbd15fa18af12d2b6de", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "83675c12902d861dc66c40b80cd7198a64328235004bd1ea7a2822ef40f6cc92", "generation_seconds_batch": 24.322730217012577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.444238305091858, "target_z": 0.992187477098356, "ce": 5.927301406860352, "genuineness": 4.6705241203308105, "identity": 0.693510115146637, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/6f1c866210bbdda9e4559a6d5ce8c4083c35d72fb6d88dbd15fa18af12d2b6de.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "6f255e64179a5ac2c10342d74a22eac942e2f3599f7046be4803a68929f716a3", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "253916406777848689f151801898f6c60d24ed4785563768933f25110241f9f2", "generation_seconds_batch": 16.858227350981906, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5906753540039062, "target_z": -0.014648467353928865, "ce": 6.078212261199951, "genuineness": 4.482243061065674, "identity": 0.7264093160629272, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/6f255e64179a5ac2c10342d74a22eac942e2f3599f7046be4803a68929f716a3.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "6fbb0b36216f8ee5f5f0d11ca55641fda910f7f93835c7fd9b0246ff2411d060", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "98b772ca51e925d630d8fa6db6eaaa30d1428f9566355f0e8c54bf35b1bec432", "generation_seconds_batch": 17.22883652697783, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.061721920967102, "target_z": 1.3281250196299805, "ce": 5.5738525390625, "genuineness": 4.833701133728027, "identity": 0.5823754668235779, "hypothesis": "He encontrado la cajita, estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/6fbb0b36216f8ee5f5f0d11ca55641fda910f7f93835c7fd9b0246ff2411d060.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "7046728f5e85b274224ece844e459f5cce11d4687e5f92079b00e815290f6ae4", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "294ec2768c5a4ebcf588650326fca1cda570a4d2d54d7fb20992b9a3d7f56f3d", "generation_seconds_batch": 22.12175811594352, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9193903207778931, "target_z": 1.4531249672833657, "ce": 6.197352409362793, "genuineness": 4.65797233581543, "identity": 0.5806978940963745, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/7046728f5e85b274224ece844e459f5cce11d4687e5f92079b00e815290f6ae4.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "706b561b910f3412a7d67463a3933e57026778aea3fa3d5e8b75c9c6c95ead11", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "85bbddf093b1f9f08fe8c4dcc5eb941ec59d4aa3330af2d3a6115b61d9882fe1", "generation_seconds_batch": 23.313514395966195, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6221528053283691, "target_z": 0.8359375425316247, "ce": 5.839932441711426, "genuineness": 4.494794845581055, "identity": 0.7285244464874268, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/706b561b910f3412a7d67463a3933e57026778aea3fa3d5e8b75c9c6c95ead11.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "70a71585f372e8bc560e1581465307eac663ff81e7845553ccdc7244df9c64cd", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2572c940ea97430de17c83319305c9a8f2c9062dd5ce572f27c133ab925c8b77", "generation_seconds_batch": 23.289206663961522, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.417551040649414, "target_z": 1.015625045803288, "ce": 5.617537498474121, "genuineness": 4.8588056564331055, "identity": 0.5882400274276733, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/70a71585f372e8bc560e1581465307eac663ff81e7845553ccdc7244df9c64cd.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "70abaf986709f961969d8b1c0da8425a8499a9be71c9574c9154205c8429d8d0", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fb009e670306afa84e385a345423ee649c315c601413d2535b3b962b971ae4b9", "generation_seconds_batch": 21.590833863941953, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.061721920967102, "target_z": 1.3281250196299805, "ce": 5.879645347595215, "genuineness": 4.406930446624756, "identity": 0.5845131874084473, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/70abaf986709f961969d8b1c0da8425a8499a9be71c9574c9154205c8429d8d0.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "70b167054ee4bf27edea7ba8bf0e62ce36b9afdf4b837bd4d009380f27c9e786", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "24667e6add4bee9323af11fd16817e39e57ff1a518746e9713d8d5d359245e25", "generation_seconds_batch": 22.747887191944756, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1417834758758545, "target_z": 1.2578125229016441, "ce": 5.605623245239258, "genuineness": 4.871357440948486, "identity": 0.7561851739883423, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise, dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/70b167054ee4bf27edea7ba8bf0e62ce36b9afdf4b837bd4d009380f27c9e786.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "70e06f175ae3dc1d6c107983a3e130c627a15b3a79ae9b0d0946b8c60db024cb", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0db5ac5db19b132f0c59e846058cf15b21f5446e2d465ad62ac3e3391c2fcdc8", "generation_seconds_batch": 23.99642169906292, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1951578855514526, "target_z": 1.2109374901850096, "ce": 5.156862258911133, "genuineness": 4.557555198669434, "identity": 0.6450973749160767, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/70e06f175ae3dc1d6c107983a3e130c627a15b3a79ae9b0d0946b8c60db024cb.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "7164ab9c4d06ee36adeaf5cf2e03c8409f7be5025446b176b41299c5214c45b8", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "43ea2157e8be16f44cbf5d40393625296297210545c9102723a69c5f7c7ebe04", "generation_seconds_batch": 16.89306024601683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3015642166137695, "target_z": 0.23925774175027817, "ce": 5.665193557739258, "genuineness": 4.268857479095459, "identity": 0.621458113193512, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/7164ab9c4d06ee36adeaf5cf2e03c8409f7be5025446b176b41299c5214c45b8.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "716b862948ee474b0b9a86ee4f3d1503d98bce1578560184c83001587e404276", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8a3022d5fc3e70479f71bac59ae79a39b11b32cbf85e5f02e3c87ea7f3944cf5", "generation_seconds_batch": 17.433165887952782, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1447768211364746, "target_z": 0.3769532092453335, "ce": 5.855817794799805, "genuineness": 4.494794845581055, "identity": 0.6669993996620178, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/716b862948ee474b0b9a86ee4f3d1503d98bce1578560184c83001587e404276.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "716d1f288aa8ef361484971c65a85a554c54575891565519bb10b3a070af12b6", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "40d54945c05adfb95b34ecaf0e658902096bac068257edde8dfedb85e60be893", "generation_seconds_batch": 17.91963773907628, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7592672109603882, "target_z": 1.5937499607400387, "ce": 5.597680568695068, "genuineness": 4.432034492492676, "identity": 0.6727123260498047, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/716d1f288aa8ef361484971c65a85a554c54575891565519bb10b3a070af12b6.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": "71c9eb8b42ed5151fb700cdd04059f031312ff26df37d822d511fe5a1db6c49e", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d00058f121ffd7a2845e1649e9de55ee634fcb7e4dfde0b6e7af2f5b9f52ef3c", "generation_seconds_batch": 16.90205033204984, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.212949275970459, "target_z": 1.1953125490749517, "ce": 5.839932441711426, "genuineness": 4.570107460021973, "identity": 0.5472099184989929, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/71c9eb8b42ed5151fb700cdd04059f031312ff26df37d822d511fe5a1db6c49e.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "71e4a84f0a92f0cefe678574a45ecd82d4e4ea7738d7c6a10b16e4a8228631a0", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4c66667950d20820be8ae7fced5b31dca1ff681b6fdf29b1df831420ef0df2f4", "generation_seconds_batch": 23.014958665007725, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.079513430595398, "target_z": 1.3124999738266925, "ce": 6.157639026641846, "genuineness": 4.519899368286133, "identity": 0.48140305280685425, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/71e4a84f0a92f0cefe678574a45ecd82d4e4ea7738d7c6a10b16e4a8228631a0.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "720a77b0775f0623aa8a13556748ba61e57311ed37b34cfdeb62262d157cebfd", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "99a9c87c7ba5b1eb2c0b32c6a6722e0b41aefa478854df621f6f8dbfdef0fa20", "generation_seconds_batch": 22.561512141954154, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5154041051864624, "target_z": 0.9296875032716634, "ce": 5.831989288330078, "genuineness": 4.545003414154053, "identity": 0.5857799649238586, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/720a77b0775f0623aa8a13556748ba61e57311ed37b34cfdeb62262d157cebfd.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "725ee77c7a6bdd947903c1dc809db4f6fa67398854e7fe82fefced1f7c741253", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e464d55a6081de4e1dfdd19e1fbcf5759bc103548e3a06224f56070a65b574ac", "generation_seconds_batch": 22.437450910103507, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7333494424819946, "target_z": 0.738281267994149, "ce": 5.927301406860352, "genuineness": 4.507347106933594, "identity": 0.6925456523895264, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/725ee77c7a6bdd947903c1dc809db4f6fa67398854e7fe82fefced1f7c741253.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "7278d598de8a45a7e3977b3256eaf3bb6c0d0c2659cfbadf74e2e882247f2d61", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "97c8a728480559ec01c8bfcab5f2bdcc48257cd6b454517733daa6a947d7d3df", "generation_seconds_batch": 23.142148748971522, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0973048210144043, "target_z": 1.2968750327166343, "ce": 5.6532793045043945, "genuineness": 4.7081804275512695, "identity": 0.7326407432556152, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise, dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/7278d598de8a45a7e3977b3256eaf3bb6c0d0c2659cfbadf74e2e882247f2d61.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "72e87c02a02814741731b37b2c09cc568f3f979e111d95d59d4095e46cf5d095", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "317de8065a77c6a77720988420dba1a0a9af6e6b9585fffa8a5296b62f88129e", "generation_seconds_batch": 21.17204432899598, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0936264991760254, "target_z": 0.42187498037001936, "ce": 6.054384231567383, "genuineness": 4.720732688903809, "identity": 0.7270908355712891, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/72e87c02a02814741731b37b2c09cc568f3f979e111d95d59d4095e46cf5d095.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": "73a79896830795c8dd7b5bb76fe09b519af8cb5159a594c5d9886e85d2178996", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9266e941e6c1753127ed671a6a5c152263e565aa2d253db9f41a0ad8ac5a5564", "generation_seconds_batch": 20.422780020046048, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3285938501358032, "target_z": 1.0937499607400387, "ce": 5.542081832885742, "genuineness": 4.6203155517578125, "identity": 0.5409533977508545, "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/vn_STNC_reg_low/73a79896830795c8dd7b5bb76fe09b519af8cb5159a594c5d9886e85d2178996.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "74105413687a5ed2df72b4ec4317ffcd087763b7b20011413a26b19ac889c498", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bed61f260ed685b6746825a96863f3d41b503ebf5068d3b59b88daeeec4ddf37", "generation_seconds_batch": 23.050241842051037, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.789717197418213, "target_z": -0.18945307837879602, "ce": 6.0305562019348145, "genuineness": 4.030367851257324, "identity": 0.41152167320251465, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/74105413687a5ed2df72b4ec4317ffcd087763b7b20011413a26b19ac889c498.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": "74591ed920d4200412adb0d0b6358b05bdd472465b80e37b3e5a5455256a68fa", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bd4d978cc4853d7deadc55495755effba12974ef063f3e097fd73f29f52d71f7", "generation_seconds_batch": 20.935181880020536, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.417551040649414, "target_z": 1.015625045803288, "ce": 5.7684478759765625, "genuineness": 4.833701133728027, "identity": 0.7985739707946777, "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/vn_STNC_reg_low/74591ed920d4200412adb0d0b6358b05bdd472465b80e37b3e5a5455256a68fa.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "7496cafefbac6a80901a68f1d71a06aedb2838d70c013066c54a7138ea2df88e", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4998a339e307f6433e40adbdacb57806af3a04bbd7d0ab1f9ad767ac689b3b58", "generation_seconds_batch": 21.8722267660778, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5154041051864624, "target_z": 0.9296875032716634, "ce": 5.696964263916016, "genuineness": 4.444586753845215, "identity": 0.47634851932525635, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/7496cafefbac6a80901a68f1d71a06aedb2838d70c013066c54a7138ea2df88e.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "74db97f403dc4688bb3af286e8600d9ed9c89b8b4e9846a14abb8dbf9d2505cf", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "53446c7477578d42aa48c66bab4b954b2fb185a5be1fd6a2495c887d9a96fa7a", "generation_seconds_batch": 17.135985934059136, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7956194877624512, "target_z": 0.6835938170691005, "ce": 5.998785495758057, "genuineness": 4.570107460021973, "identity": 0.5148847103118896, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/74db97f403dc4688bb3af286e8600d9ed9c89b8b4e9846a14abb8dbf9d2505cf.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "75181118fa641add7934264fa62a510d4d426d68495161a5bd1514de8ec01ea1", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f98b8548a889c9fc9ce41ffa8245b5aa07ed4719a28b8b4fd9527b859f0d6485", "generation_seconds_batch": 21.313703596009873, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.942399024963379, "target_z": 0.5546875556182784, "ce": 5.772419452667236, "genuineness": 4.607763767242432, "identity": 0.749455988407135, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/75181118fa641add7934264fa62a510d4d426d68495161a5bd1514de8ec01ea1.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "75563696197cd5f130e97b53d6368b41509c887ecba2fb10f68685c0d73469f0", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4216ed3439ef63c19e0255383d805b435d69f1e5b59d69d1682cfadb00e2912e", "generation_seconds_batch": 15.933410729980096, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.779709577560425, "target_z": -0.1806640817210227, "ce": 6.09409761428833, "genuineness": 4.394378662109375, "identity": 0.32512620091438293, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/75563696197cd5f130e97b53d6368b41509c887ecba2fb10f68685c0d73469f0.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "75752318cf3b1d933d8481b4b1d13adbc6871415c7424615c5311a5e74bb097d", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "42770452f1a063d5ef9db1ef81065de1bc9c1fce2c8d397e4f2bae792f0bed55", "generation_seconds_batch": 25.52765471499879, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6651771068573, "target_z": -0.08007817652869914, "ce": 6.141753673553467, "genuineness": 4.444586753845215, "identity": 0.7015774250030518, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/75752318cf3b1d933d8481b4b1d13adbc6871415c7424615c5311a5e74bb097d.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "75a59249781c80ba9c55af81bf5e92621dd5d7e205b7dba16447b36e27404e17", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e356281807f33794b5c35c930aa98c1c63addd61401464e58167cdc9c7ee9868", "generation_seconds_batch": 23.269974588998593, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.61770498752594, "target_z": 0.8398437516358317, "ce": 5.974957466125488, "genuineness": 4.300237655639648, "identity": 0.3955618143081665, "hypothesis": "The 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/vn_STNC_reg_low/75a59249781c80ba9c55af81bf5e92621dd5d7e205b7dba16447b36e27404e17.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "75e900b9479e9a4f51021123ca5d50bbb653392dc5e9cadd79a1f624dcc3a08f", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "6bdf2476f5464da53006ef3a6067921b723e936cf1f29b0b9ea9bc303f7ed217", "generation_seconds_batch": 17.41999003395904, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6666314601898193, "target_z": 0.7968750327166344, "ce": 5.677107334136963, "genuineness": 4.519899368286133, "identity": 0.7036371231079102, "hypothesis": "Die Tür 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/vn_STNC_reg_low/75e900b9479e9a4f51021123ca5d50bbb653392dc5e9cadd79a1f624dcc3a08f.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "760b94d6d530454d671203824dd98970543d0ab6313708c6e6bf84e032f10839", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3941bf0dd1d8e13c083d5aec203c39a807a4244450282ea1be445e4336ae4d29", "generation_seconds_batch": 17.214738643961027, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6184744834899902, "target_z": -0.039062509814990314, "ce": 6.1894097328186035, "genuineness": 4.507347106933594, "identity": 0.6457496881484985, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/760b94d6d530454d671203824dd98970543d0ab6313708c6e6bf84e032f10839.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "76308ccd06be439f5c90c1faf65679a97d0a2cb98e55d71fe68e601e1f62cb38", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "32db388347841d7ad15a266666289e5c569f4ded230ec313ec546db8836807b3", "generation_seconds_batch": 17.273282513022423, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1862621307373047, "target_z": 1.2187500130866538, "ce": 6.078212261199951, "genuineness": 4.532451152801514, "identity": 0.319636732339859, "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/vn_STNC_reg_low/76308ccd06be439f5c90c1faf65679a97d0a2cb98e55d71fe68e601e1f62cb38.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "7691d8a5600671e7786226ca643c2a0b43a0123c0f37075df44bcae59c45bcf5", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2d93553c919ceac2854d285ea1fdcf1f3abcee8b7f3b16454ef6eaad95cb9e5e", "generation_seconds_batch": 21.430985065060668, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.487262487411499, "target_z": 0.07617186273126211, "ce": 6.117925643920898, "genuineness": 4.20609712600708, "identity": 0.4688069224357605, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/7691d8a5600671e7786226ca643c2a0b43a0123c0f37075df44bcae59c45bcf5.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "769251b554af1e31f6093590c3d42659b15c0d67a848025abf74bf7e94a2d811", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "258d6fdd72153082e8c8113e57e37aa5811dc0382b02caf8a4f2fba8beef9da6", "generation_seconds_batch": 23.105471969000064, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.313795804977417, "target_z": 0.22851558819378634, "ce": 6.1894097328186035, "genuineness": 4.532451152801514, "identity": 0.6981043219566345, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/769251b554af1e31f6093590c3d42659b15c0d67a848025abf74bf7e94a2d811.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "771e6ce4078f28a3d569e1b5830d9eb4f4da3cf1f8a865937114394c680168da", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e01b4d0eae5cdc95ec307c0f3b0ce598c2a79e5b0ccf7f08f5d7c2a9990d7ea2", "generation_seconds_batch": 17.34863237803802, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3374894857406616, "target_z": 1.0859375425316247, "ce": 5.442798614501953, "genuineness": 4.6203155517578125, "identity": 0.5718333125114441, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/771e6ce4078f28a3d569e1b5830d9eb4f4da3cf1f8a865937114394c680168da.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "775290e2d0b77bd7261380b6315161930e49d02b4ddb5e089792defe10ff2633", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "daad29bc254e5dcb14cab2cefba3d6b298dbad0ec38745ca4c98417be623fcc9", "generation_seconds_batch": 22.85736859298777, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8660159111022949, "target_z": 1.5, "ce": 5.7128496170043945, "genuineness": 4.243753433227539, "identity": 0.6710021495819092, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/775290e2d0b77bd7261380b6315161930e49d02b4ddb5e089792defe10ff2633.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "77603bdb9d5ca358f95e4d60d8d13d7407e73faade1dc09ed1515c76420be1cd", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "010c5b04a4e4f10db2d4498fd2b618c88271e269ae1d32bebc4c27983a8fc0c6", "generation_seconds_batch": 22.530962995020673, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1150963306427002, "target_z": 1.2812499869133462, "ce": 5.824047088623047, "genuineness": 4.6956281661987305, "identity": 0.7454422116279602, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/77603bdb9d5ca358f95e4d60d8d13d7407e73faade1dc09ed1515c76420be1cd.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "778c70db01b108d22594b40f45234dc9e484c196b325e9fa40669ce1c51197ca", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5f7aae36f5c14efe7c7e68c69544836bcb286d8b0f61f7ca2c816b58a5be44c2", "generation_seconds_batch": 16.60807763005141, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.297116279602051, "target_z": 0.2431640555477152, "ce": 6.133810997009277, "genuineness": 4.413206577301025, "identity": 0.7302788496017456, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/778c70db01b108d22594b40f45234dc9e484c196b325e9fa40669ce1c51197ca.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "7a6e3a30e6551be8ba0e75cc9a11fdd2263f320854b64da698700e3d8d6bf602", "generation_seconds_batch": 16.350541777093895, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.664621114730835, "target_z": -0.0795898873040195, "ce": 6.078212261199951, "genuineness": 4.68307638168335, "identity": 0.8146524429321289, "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/vn_STNC_reg_low/77962898fb95ddc0e92a153772191fe01482e40f98d2245158cb92652e8a93c1.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "78eddd3c4d7edd5072f221ddf9529b9c52f9ffe718bda31bb4d817a02f8103c1", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "16f6aaa25c098d7e1aed9f13e28dbd97068f4d247d54d8f924ba5ed2643f0e9d", "generation_seconds_batch": 22.527320639928803, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.578443765640259, "target_z": -0.003906313797437026, "ce": 5.839932441711426, "genuineness": 4.2374773025512695, "identity": 0.6231233477592468, "hypothesis": "Ik heb het een kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/78eddd3c4d7edd5072f221ddf9529b9c52f9ffe718bda31bb4d817a02f8103c1.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "7938f7d3a4dbf773e3dad1dd3876c74768d6759ef58bdbe47b8bf95bb6cbdc30", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1b8edb7d16120a571daa9d08b7fdda1090338a8c52d4bd78900e6ddade643b85", "generation_seconds_batch": 18.236230348004028, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9298250675201416, "target_z": -0.3125000785199225, "ce": 6.38797664642334, "genuineness": 4.243753433227539, "identity": 0.5919029712677002, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/7938f7d3a4dbf773e3dad1dd3876c74768d6759ef58bdbe47b8bf95bb6cbdc30.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "04d3b880e77aa4ae52eb4496e17414c6113b8c1dbd9d94c25e7105760f239b24", "generation_seconds_batch": 18.105520284036174, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.032125949859619, "target_z": -0.4023438301557542, "ce": 6.062326908111572, "genuineness": 4.532451152801514, "identity": 0.6910628080368042, "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/vn_STNC_reg_low/794d3b9296e3c4be7c1e4182a2fa3cf095f7d593990763ce0f19c1f5e499a989.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "798e265060cb542451b65e14d8b32476a8f56df708ff92a44fa215c8961c97cb", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0b93ef9699795dc386ec0eb5403ea86e88339796919a235e200254597ab864d6", "generation_seconds_batch": 22.297051931964234, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6703099012374878, "target_z": 1.6718749803700195, "ce": 5.919358730316162, "genuineness": 4.457139015197754, "identity": 0.6010271310806274, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/798e265060cb542451b65e14d8b32476a8f56df708ff92a44fa215c8961c97cb.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "79deb6fc38d46d9beb56e8d8c7e9b8a2b08a05ec61b6f929fde1f82451a18107", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "90b77ca1f481711efaf138e52de11c89e68a505eec0ad15575ad4f34952815ad", "generation_seconds_batch": 22.718832734040916, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9460774660110474, "target_z": 1.4296875032716634, "ce": 5.903473377227783, "genuineness": 4.6705241203308105, "identity": 0.3288399279117584, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/79deb6fc38d46d9beb56e8d8c7e9b8a2b08a05ec61b6f929fde1f82451a18107.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "7a227ecd81610a997b9268faf79a01e03446d59d006e949d46d0eb94c3a285fa", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "31acb8ee5994dc7d2c05310bc2bf1414dd7f3b691867e14749c490a5b6e48c86", "generation_seconds_batch": 20.553295812918805, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3463852405548096, "target_z": 1.0781250196299805, "ce": 6.022613525390625, "genuineness": 4.645420074462891, "identity": 0.6569849252700806, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/7a227ecd81610a997b9268faf79a01e03446d59d006e949d46d0eb94c3a285fa.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "7a6b2b6c043528056a9c317d03c4b3aec77b1f9e2c9069b93526d6fee0a68686", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d6076fe55e22d437a8c133620619a97d20b540b89af5ef469ad5c12927738be9", "generation_seconds_batch": 21.284522460075095, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7555887699127197, "target_z": 0.7187500130866538, "ce": 6.0146708488464355, "genuineness": 4.68307638168335, "identity": 0.7450315952301025, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/7a6b2b6c043528056a9c317d03c4b3aec77b1f9e2c9069b93526d6fee0a68686.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "7a77ff31540a8df1213af90e0924e1d8f281a590eed75e429a9609a30b35c209", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d9218ce2018b3a40f28c82fecbde678e06daf4799276813c3335f6d15ea80a2d", "generation_seconds_batch": 17.208157390006818, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7733802795410156, "target_z": 0.7031249672833656, "ce": 5.657250881195068, "genuineness": 4.030367851257324, "identity": 0.3491046726703644, "hypothesis": "The 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/vn_STNC_reg_low/7a77ff31540a8df1213af90e0924e1d8f281a590eed75e429a9609a30b35c209.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "7a91919cc22b276dc09e5ee383022830d978c231fda2534c117273c8c8605724", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a08c360d2a5119fe1bd9ba7e521b261d8afbaab3814dee3ddea58d764f046552", "generation_seconds_batch": 16.459219877957366, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8052847385406494, "target_z": -0.20312496728336563, "ce": 6.340320587158203, "genuineness": 4.218649387359619, "identity": 0.15983828902244568, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/7a91919cc22b276dc09e5ee383022830d978c231fda2534c117273c8c8605724.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "7a9b62d8191bb87f982ac7c32bb6c5275c8af55eaa56df33ede6bf6e92be2a87", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "510b8f44e6d752199a29d1f874bfe7686d2e67a4d244b0e3c474f0aa2c451cef", "generation_seconds_batch": 22.899351377040148, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2574279308319092, "target_z": 1.1562500392599613, "ce": 5.935244083404541, "genuineness": 4.482243061065674, "identity": 0.7472072243690491, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/7a9b62d8191bb87f982ac7c32bb6c5275c8af55eaa56df33ede6bf6e92be2a87.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "7b6fff7bb609c6c94c4eb7d08286201a207fd84fcf148f8e494d01a5efb69e46", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "52a0a985e5b68f88fd21f537246b9af73026792081e5e15851a5eb804affac29", "generation_seconds_batch": 27.057254050974734, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.942399024963379, "target_z": 0.5546875556182784, "ce": 5.760505676269531, "genuineness": 4.570107460021973, "identity": 0.6547095775604248, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/7b6fff7bb609c6c94c4eb7d08286201a207fd84fcf148f8e494d01a5efb69e46.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "7bdb25fc7087d06f19c5e22982999f546e78f5934447a0910aef3fd26bb8571a", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7abaf20768d07bb823fb7a089be5e665de267539fa78ef8d702d15a650742a5a", "generation_seconds_batch": 23.10092718701344, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.240406036376953, "target_z": 0.29296871891919735, "ce": 6.348263263702393, "genuineness": 4.68307638168335, "identity": 0.39134928584098816, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/7bdb25fc7087d06f19c5e22982999f546e78f5934447a0910aef3fd26bb8571a.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "7c2eb14b12a184142b7b3fcc4245936ada29bb439f0bf397b8bf5d6834345f0f", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a5c306de67fa9607d4869736d8f6cf78d790edd90415b70b7e8a1c9d1003e0aa", "generation_seconds_batch": 16.752372315968387, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8801289796829224, "target_z": 0.6093750065433269, "ce": 6.0305562019348145, "genuineness": 4.469690799713135, "identity": 0.6705228090286255, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/7c2eb14b12a184142b7b3fcc4245936ada29bb439f0bf397b8bf5d6834345f0f.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "7c364e0732eaf11190865fe03e51d8a476facbc9414f404daadddf46c032a4bd", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dd7212e1632a29bb77596218faf6ab0d5f519dfd823fff85d35a08a9398e1b79", "generation_seconds_batch": 22.824129321961664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.061721920967102, "target_z": 1.3281250196299805, "ce": 5.8875885009765625, "genuineness": 4.406930446624756, "identity": 0.6941722631454468, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/7c364e0732eaf11190865fe03e51d8a476facbc9414f404daadddf46c032a4bd.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "7c394803b141b4d9bbe615f4c6d7aea5e89ca978fe2812ae467bf3903d08513a", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "985653079e076c4ae6b2d3463299463c060760956b90037d7afc52c4ec07b7a0", "generation_seconds_batch": 16.414698014967144, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2604212760925293, "target_z": 0.2753907256036507, "ce": 6.1894097328186035, "genuineness": 4.482243061065674, "identity": 0.5532574653625488, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/7c394803b141b4d9bbe615f4c6d7aea5e89ca978fe2812ae467bf3903d08513a.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "7c43393ec00ae806aeaee538630ddd003edb06bc99b747c1b4cf8b3858146689", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6b5e6b11821d5cae73179edbe3939d82f318389b11ea89637f1059c562a8e22e", "generation_seconds_batch": 16.89306024601683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6043614149093628, "target_z": 0.8515624836416829, "ce": 5.454712867736816, "genuineness": 4.632867813110352, "identity": 0.8420336842536926, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/7c43393ec00ae806aeaee538630ddd003edb06bc99b747c1b4cf8b3858146689.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "7c4c13d1d3c7e3c314e939edf6fbb7312930e3d60afde2b4db4f56d433ad97e0", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "26785ceb69e1992dad4e1e18cd1b253f753c24c7dc153fd501e89acfac9060ca", "generation_seconds_batch": 21.349195611081086, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9157118797302246, "target_z": 0.5781250196299806, "ce": 5.982900142669678, "genuineness": 4.582659721374512, "identity": 0.7645604014396667, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/7c4c13d1d3c7e3c314e939edf6fbb7312930e3d60afde2b4db4f56d433ad97e0.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "7caf7cd3175eb9f300c86c33360cbe965634340cf71f1996db2da3da171d671c", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d0f9eb884bfa35b2181db70a889bab35fc8c69eb1d15472ddce0535254795381", "generation_seconds_batch": 16.86311360099353, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3571624755859375, "target_z": 0.19042965682815527, "ce": 6.09409761428833, "genuineness": 4.846253395080566, "identity": 0.5149577856063843, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/7caf7cd3175eb9f300c86c33360cbe965634340cf71f1996db2da3da171d671c.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "7d2728a369029b04954c1aecf29e5d9881dd1c00466dfdb48cc7c5a5dc90772f", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4e6328c2040939522ba7d76886f8a1241c93820f4bab9f1375b4587f40880b09", "generation_seconds_batch": 22.851364523055963, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.079513430595398, "target_z": 1.3124999738266925, "ce": 5.629451274871826, "genuineness": 4.68307638168335, "identity": 0.5954320430755615, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prête.", "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/vn_STNC_reg_low/7d2728a369029b04954c1aecf29e5d9881dd1c00466dfdb48cc7c5a5dc90772f.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "7e2689a5eeca6b51a91cfd52ce439b3d6b64898719f7012eca550a45e8da59f6", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5569e74d275e7c5ddbb8653bce40763e50d0c03044bd508a7581c49393424761", "generation_seconds_batch": 17.87239120702725, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5687785148620605, "target_z": 0.8828124705550291, "ce": 5.863759994506836, "genuineness": 4.645420074462891, "identity": 0.7837046384811401, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/7e2689a5eeca6b51a91cfd52ce439b3d6b64898719f7012eca550a45e8da59f6.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "7e26a51531799a34cac9bf4a7fea661185d80eddb201490892b82b911798efa7", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4df82c362ededbeef498bc9ba392f9f5bed0d9109281b954c303304fbb56758d", "generation_seconds_batch": 22.50628304295242, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7022144794464111, "target_z": 0.7656249411100581, "ce": 6.125868320465088, "genuineness": 4.758388519287109, "identity": 0.6509801149368286, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/7e26a51531799a34cac9bf4a7fea661185d80eddb201490892b82b911798efa7.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "7e87dbc98e8305c20e5927243668da1182e86e9b9a28ebb4d8a3be6f0f86c03d", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1c78d0beddeb17f1e0cef6a2f95e3e5bc6123bdb72e26df7c6fd44810624390f", "generation_seconds_batch": 17.11507529800292, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6868603229522705, "target_z": -0.09912103751828467, "ce": 6.221180438995361, "genuineness": 4.419482707977295, "identity": 0.5282876491546631, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/7e87dbc98e8305c20e5927243668da1182e86e9b9a28ebb4d8a3be6f0f86c03d.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "7ee82d7fd7a376737acf3d87e650ed8f0be77ab12164df4696599608f36359e1", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "abca344e75af968371b6d930fb16e9cc76078b10d030d3ddab45cf4588a2e0f3", "generation_seconds_batch": 22.832411274896003, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7948501110076904, "target_z": 1.5624999738266925, "ce": 5.426913261413574, "genuineness": 4.7081804275512695, "identity": 0.5364593267440796, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/7ee82d7fd7a376737acf3d87e650ed8f0be77ab12164df4696599608f36359e1.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "7f46c90120561f32e78b1bd2e5667d1dce9950141dce8b9e8251cf456965ee36", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e7c8e1928b6745bef98b4c4dab3df219cefc93683ada71c9c6af8471a540cfa6", "generation_seconds_batch": 24.039401449030265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1062005758285522, "target_z": 1.2890625098149904, "ce": 5.681078910827637, "genuineness": 4.645420074462891, "identity": 0.5278120040893555, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/7f46c90120561f32e78b1bd2e5667d1dce9950141dce8b9e8251cf456965ee36.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "7f4b0f62db854ce128ab694fab66184044d230c9d350182ce96e24e7aaee2be7", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "075591b873085dba61ed85aaa20012020c5f0c197b503a419ccdf6670a990716", "generation_seconds_batch": 23.271698817028664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7859543561935425, "target_z": 1.5703124967283366, "ce": 5.6334228515625, "genuineness": 4.607763767242432, "identity": 0.6511935591697693, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/7f4b0f62db854ce128ab694fab66184044d230c9d350182ce96e24e7aaee2be7.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "eaaacb93f4c33822f97438ad9a5ba9632bdaf49b08d7fc4782adf5b4324508b7", "generation_seconds_batch": 16.299843924934976, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3360350131988525, "target_z": 0.20898443797952115, "ce": 6.300607204437256, "genuineness": 4.325342178344727, "identity": 0.3309076428413391, "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/vn_STNC_reg_low/7f6fce1a6f733719f45a363f28fee4e24b2f07a3396aa07b42392913d7c24239.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "7fb92a87be6f6e0a7d167e7d0d7f97d68ba526431072c24b8248874448c3c62b", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fdf8b2034d74848adb848b5ce95539d1551e78716acd85ee357c0285ce3ad300", "generation_seconds_batch": 22.543397480039857, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7111101150512695, "target_z": 0.757812522901644, "ce": 6.1020402908325195, "genuineness": 4.65797233581543, "identity": 0.5729904770851135, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/7fb92a87be6f6e0a7d167e7d0d7f97d68ba526431072c24b8248874448c3c62b.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "801b9a54ec8cc30d5c9a955f144ea7c1766ea326bf613c4441373eadee1605e3", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2c65681017e42de7f3e19e8d7b4c42d802ba43f0247c047080106bffe9710f62", "generation_seconds_batch": 22.89835396001581, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.80006742477417, "target_z": 0.6796875032716634, "ce": 6.165581703186035, "genuineness": 4.507347106933594, "identity": 0.6253023147583008, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/801b9a54ec8cc30d5c9a955f144ea7c1766ea326bf613c4441373eadee1605e3.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "801c3d79afa8fbae0e7bc2742357e122fe74c023a60446b6c89aa4ba6d828045", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d7cacdb503cce728f8df7dd541975cb6d892e2ae35f77147a760698d8ba20db9", "generation_seconds_batch": 16.414698014967144, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5689918994903564, "target_z": 0.004394603022116654, "ce": 6.348263263702393, "genuineness": 4.432034492492676, "identity": 0.3920794725418091, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/801c3d79afa8fbae0e7bc2742357e122fe74c023a60446b6c89aa4ba6d828045.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "80396e68f5d602fa1323872474bb1ec92cdf01906a3a24fa511d134afa7853a6", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1e44d1a06678f812327576fabdad5be4db57e001c389920aa5a568347f3fbe8e", "generation_seconds_batch": 15.933410729980096, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.601238965988159, "target_z": -0.023925753236381818, "ce": 5.982900142669678, "genuineness": 4.250029563903809, "identity": 0.31612610816955566, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/80396e68f5d602fa1323872474bb1ec92cdf01906a3a24fa511d134afa7853a6.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "4c07abb1dc701721a36e3dd3350b9c335cc658d23c4bb94be87969f3096aaa24", "generation_seconds_batch": 17.569804698927328, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0832762718200684, "target_z": -0.4472656012804401, "ce": 6.252951145172119, "genuineness": 4.394378662109375, "identity": 0.7032127380371094, "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/vn_STNC_reg_low/80c2597a6b8f43b5338a6b37d654868d679f46ded48514b42837ab18fba327ad.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "80fb7a32c381a646f5ca960ec0d0132188b40899eda580b0a6921e6f40aaae77", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "153bad4dc2716e796571b11b236e96d8dcada7ed556058b23222d3bbc0d0427a", "generation_seconds_batch": 23.34666066803038, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.184807777404785, "target_z": 0.3417968038413202, "ce": 5.7247633934021, "genuineness": 4.444586753845215, "identity": 0.6839971542358398, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/80fb7a32c381a646f5ca960ec0d0132188b40899eda580b0a6921e6f40aaae77.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "81cf44d1aa3f9d2ce108d06ec959a54793c371069eff8dc2e35012b2d55137f0", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3f7bf5d506370c4f7788a35069d7d51995cd47646db3c796074888aa4f25320f", "generation_seconds_batch": 16.41353548504412, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.244853973388672, "target_z": 0.28906240512176035, "ce": 6.38797664642334, "genuineness": 4.319066047668457, "identity": 0.3954746723175049, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/81cf44d1aa3f9d2ce108d06ec959a54793c371069eff8dc2e35012b2d55137f0.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "829f71676a914d615fa9b7dc3319ac4038771c746139501d810517ee8c9ba5b3", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "83e4d18071638dbdc6f59a5bfe0866a61eca2691ed2466dfd321936b8d1a52bf", "generation_seconds_batch": 16.4634020839585, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3960812091827393, "target_z": 0.15625003925996125, "ce": 6.0305562019348145, "genuineness": 4.7081804275512695, "identity": 0.7750325798988342, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/829f71676a914d615fa9b7dc3319ac4038771c746139501d810517ee8c9ba5b3.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "82df89094a02edf2a43e41aca1025f820edd58f454054bbd9e1ee0023ee012ce", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a83b1019b4ef01af1da61e6c43d26072e4644af66690796f92ae86003e4ec55f", "generation_seconds_batch": 17.03882373892702, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4405598640441895, "target_z": 0.11718752944497093, "ce": 6.300607204437256, "genuineness": 4.632867813110352, "identity": 0.5481240153312683, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/82df89094a02edf2a43e41aca1025f820edd58f454054bbd9e1ee0023ee012ce.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "82ed3d4bfdd05ad6a98c67475ac8aef8bbed8ce7f3865336038550272bb1d9bb", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b1e5f71d2d6257b113a4b7a952979fce4c57daadb621d32cca6c1387520b95fe", "generation_seconds_batch": 22.038143103942275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1328877210617065, "target_z": 1.265625045803288, "ce": 5.601652145385742, "genuineness": 4.582659721374512, "identity": 0.7943907976150513, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/82ed3d4bfdd05ad6a98c67475ac8aef8bbed8ce7f3865336038550272bb1d9bb.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": "8310e7a67f473bad0022076fcee24a28df42a0299a51059f813af3aa617fa2dc", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9896b53c0fe0f50a3204ceda01a24de5f45ba7e1acd8cf89c4b4c5553c623e9d", "generation_seconds_batch": 16.662871649954468, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9371817111968994, "target_z": 1.4375000261733075, "ce": 5.780362129211426, "genuineness": 4.482243061065674, "identity": 0.2529352307319641, "hypothesis": "De weg gaat linksaf naar de brug, loopt verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.09523809523809523, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/8310e7a67f473bad0022076fcee24a28df42a0299a51059f813af3aa617fa2dc.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "835f663983fed720dc60fec96c2f245ec7dee378af163b3c6809515e9829ac86", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d4184c97a773f32b62233b41dc6fc9b116182bcf6fcc3dc3d8a18251c61e06d7", "generation_seconds_batch": 21.561668646987528, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9335033893585205, "target_z": 0.5624999738266925, "ce": 5.927301406860352, "genuineness": 4.6956281661987305, "identity": 0.6955162882804871, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/835f663983fed720dc60fec96c2f245ec7dee378af163b3c6809515e9829ac86.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "837c595d70dd318483f7bee92c964394857b4a17861300b3b8a5930a6ffb0360", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "45e81b725a1511cf431c79a30a60a7b2a0ff151f59cc12201c2eeb57fc7c357a", "generation_seconds_batch": 17.007239979924634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7422451972961426, "target_z": 0.7304687450925048, "ce": 6.0146708488464355, "genuineness": 4.720732688903809, "identity": 0.7592854499816895, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/837c595d70dd318483f7bee92c964394857b4a17861300b3b8a5930a6ffb0360.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "83c65b5d9d373ef967f29f725b34d787e8868735bfb021797b7e557a7840d25c", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bc6593541d9bb1a02b748e76a6a6d44a3e82f68eda2aa0dfc48798fc9f4c961d", "generation_seconds_batch": 21.523821343900636, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5584282875061035, "target_z": 0.013671888904569609, "ce": 5.982900142669678, "genuineness": 4.632867813110352, "identity": 0.6744401454925537, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/83c65b5d9d373ef967f29f725b34d787e8868735bfb021797b7e557a7840d25c.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "83ddfebd1bb93e0b1a9e503ac9e8429d70b0386f82e9237d2f0a6fffdea5ac3a", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "667ab8201bbd245c2b8d509f5b8527a47aee8d0a063bc783e146398a02700e4a", "generation_seconds_batch": 16.557510982034728, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9201598167419434, "target_z": 0.5742187058325436, "ce": 6.062326908111572, "genuineness": 4.632867813110352, "identity": 0.6348531246185303, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/83ddfebd1bb93e0b1a9e503ac9e8429d70b0386f82e9237d2f0a6fffdea5ac3a.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "83f0a7dc5c38c4923081f92c8adbc8e587da2caf4d09ace01594bace09b1df8c", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6ecf81dcf75c35183df909354cd69a09f888d110d3837473244e6ce4da77df32", "generation_seconds_batch": 21.185890013934113, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.102522134780884, "target_z": 0.4140625621616053, "ce": 6.109982967376709, "genuineness": 4.595211505889893, "identity": 0.6552858352661133, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand, toi prêt?", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/83f0a7dc5c38c4923081f92c8adbc8e587da2caf4d09ace01594bace09b1df8c.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "84338c4e27043cba756efc72b3346311e2ebf8975c835c5a6b2cd2b344e1ed55", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "8dc14f37d4eddff62877f345efe21d2f564431abdf65cc4717b7ab9ce2637ba5", "generation_seconds_batch": 16.40111226902809, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.275219440460205, "target_z": 1.1406249934566732, "ce": 5.903473377227783, "genuineness": 4.607763767242432, "identity": 0.6059147119522095, "hypothesis": "The 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/vn_STNC_reg_low/84338c4e27043cba756efc72b3346311e2ebf8975c835c5a6b2cd2b344e1ed55.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": "8452b6aede53d7335bf19a663ad5cb2784a24abcc548ddecaffa30e4bd183009", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "1e591a43a5a882f6450bddcdbf6c1e55f1395a99cc84f5f7811c7bfa3890c283", "generation_seconds_batch": 16.98819609102793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4931647777557373, "target_z": 0.9492187581791586, "ce": 5.935244083404541, "genuineness": 4.632867813110352, "identity": 0.6299319267272949, "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": 0.120000000000001, "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/vn_STNC_reg_low/8452b6aede53d7335bf19a663ad5cb2784a24abcc548ddecaffa30e4bd183009.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "84a0a3c2bbe88302efb4235442462f02b695f032758e78478824bce8f64516c4", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "101e5fecf1685373976e44ecf8c70f4343a70147d2f92a00e8d8e4c821789702", "generation_seconds_batch": 18.9860270350473, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2493016719818115, "target_z": 0.2851563007107833, "ce": 6.356205940246582, "genuineness": 4.570107460021973, "identity": 0.43052032589912415, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/84a0a3c2bbe88302efb4235442462f02b695f032758e78478824bce8f64516c4.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "84da605e98138a99de01c13a100dca4ac7dc5e5c1bb8bb1f7411f6b168deb64b", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7c67fafea4cbb37e3d4768726913b89dd8e166f0f77ae49f532943a8cd949962", "generation_seconds_batch": 21.74014264100697, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5331956148147583, "target_z": 0.9140624574683753, "ce": 5.661221981048584, "genuineness": 4.796044826507568, "identity": 0.5567020177841187, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/84da605e98138a99de01c13a100dca4ac7dc5e5c1bb8bb1f7411f6b168deb64b.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "84de2133642b2f63f466219107287955041c9eabae6c877e1b0dc1c2f3747d99", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "6b34f5531c6755669d6b0ce8bce51b62af15b7e44e794225b141592e2ba90b7f", "generation_seconds_batch": 22.75139091105666, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1150963306427002, "target_z": 1.2812499869133462, "ce": 5.744620323181152, "genuineness": 4.607763767242432, "identity": 0.7772536277770996, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/84de2133642b2f63f466219107287955041c9eabae6c877e1b0dc1c2f3747d99.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "853542f992c3109e8e1d0aeae53a4278319c2e294d912b0d8d86a35aee9cdf81", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ad5c66380e94196a43af1b7e8091f90d315d8127d1b171738ed05f67483e257c", "generation_seconds_batch": 16.60807763005141, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.858659267425537, "target_z": -0.25000010469323, "ce": 6.205295085906982, "genuineness": 4.482243061065674, "identity": 0.626702070236206, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/853542f992c3109e8e1d0aeae53a4278319c2e294d912b0d8d86a35aee9cdf81.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "8536f84df8afe5972c5a8ba685de02e12efa7945497bd5e25a4ccbaab5552437", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "40ceccf1e95a8dea21983351b90277f6098119b9fe258ec32adf2b862801d0e1", "generation_seconds_batch": 22.096640227013268, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3908638954162598, "target_z": 1.0390625098149904, "ce": 5.911416053771973, "genuineness": 4.545003414154053, "identity": 0.35824257135391235, "hypothesis": "The door's still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/8536f84df8afe5972c5a8ba685de02e12efa7945497bd5e25a4ccbaab5552437.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "86367a4273d60aacef3e556fa6deab3b7e25a49b69ce63d9491ff59468dc8b6b", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a5464d207d892ad70ede532c6896e01e3d69719cafbc2a56f55be567b0a1b443", "generation_seconds_batch": 22.33053092902992, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.608809232711792, "target_z": 0.8476562745374758, "ce": 5.911416053771973, "genuineness": 4.821148872375488, "identity": 0.7763240337371826, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/86367a4273d60aacef3e556fa6deab3b7e25a49b69ce63d9491ff59468dc8b6b.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "8672943b4b8802ade86041e2a5f3abeddbe1b70703788aed5b519981e0dfe473", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "83e1774c642e457becef3cd0e32578986c6d1cb794db53629e8deab75aea6972", "generation_seconds_batch": 18.9860270350473, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1114180088043213, "target_z": 0.40624993456673125, "ce": 6.324435234069824, "genuineness": 4.6705241203308105, "identity": 0.4135306477546692, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/8672943b4b8802ade86041e2a5f3abeddbe1b70703788aed5b519981e0dfe473.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "8697d4050730564e086a23e3fc2ca6a3e7cd05c996dba8b27d2b30e5bb835547", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e916ca978dc4a597d4b22813a329250bfb75aa5200b7ce2b3de03492cfd9e6aa", "generation_seconds_batch": 22.923548455932178, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8801289796829224, "target_z": 0.6093750065433269, "ce": 5.784333229064941, "genuineness": 4.017816066741943, "identity": 0.25563323497772217, "hypothesis": "The door i is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/8697d4050730564e086a23e3fc2ca6a3e7cd05c996dba8b27d2b30e5bb835547.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "86b006934cedcd4875836d067d1da27ef02944ac3229dac075cb7735908756c5", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7d0f0b966010dc895f25acf2132d1f8d3439c9a9e1e2734813321304bdb45f04", "generation_seconds_batch": 17.87239120702725, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7022144794464111, "target_z": 0.7656249411100581, "ce": 5.550024509429932, "genuineness": 4.394378662109375, "identity": 0.8017603158950806, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/86b006934cedcd4875836d067d1da27ef02944ac3229dac075cb7735908756c5.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": "86c3162c0e4f4d1ec0c50c3952fb40d204391beebf9a9a896ef49f1b6611c5f2", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ae1edebb35f89cb4b0abf155ced26657752432657ddc4fcfd99a117f6f7b40e8", "generation_seconds_batch": 20.57339324709028, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4976125955581665, "target_z": 0.9453125490749515, "ce": 5.546053409576416, "genuineness": 4.632867813110352, "identity": 0.650477409362793, "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/vn_STNC_reg_low/86c3162c0e4f4d1ec0c50c3952fb40d204391beebf9a9a896ef49f1b6611c5f2.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "86d35476242bdae3806089dc772d2c7a2bec5c1f557142984822a38447f69c0c", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3b70dfacbda155e596f844cf3cdfc03f5fbf44080e22079f8ac706a29621e613", "generation_seconds_batch": 15.868076462997124, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.328251361846924, "target_z": 0.21582027773857598, "ce": 6.109982967376709, "genuineness": 4.7332844734191895, "identity": 0.5900629758834839, "hypothesis": "Ik heb het kleine nociau gevonden. Het stond naast dit oude raam. We kunnen er samen in kijken.", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/86d35476242bdae3806089dc772d2c7a2bec5c1f557142984822a38447f69c0c.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "86e05533cefa1a9e29c15574766a384414573a757cf913d3d5660d1f0caf3a9c", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7eca2658d4dbdf7ce3f1962dc75d5fb57cd7a463bae261e04317fbcf123f2ea5", "generation_seconds_batch": 16.183236254029907, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.629594087600708, "target_z": -0.04882808492212289, "ce": 6.09409761428833, "genuineness": 4.632867813110352, "identity": 0.49139365553855896, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/86e05533cefa1a9e29c15574766a384414573a757cf913d3d5660d1f0caf3a9c.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "86e8a94fd3f6ed44e785d322058e3533772d32baf2be12c3ccedca5e9b936199", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dd240aa3e147ca75a02bd456d64ee34a085cc7d2765d74ea82913aa154f2328c", "generation_seconds_batch": 16.72890748793725, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.591017723083496, "target_z": 0.8632813203407639, "ce": 6.0146708488464355, "genuineness": 4.419482707977295, "identity": 0.3317118287086487, "hypothesis": "The door is still open. I I put your coat on the chair. Tell me when you are ready.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/86e8a94fd3f6ed44e785d322058e3533772d32baf2be12c3ccedca5e9b936199.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": "87ed21416792ca114796bc43f92d889b108f2ba2874f73eafd4c2040d327ec42", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5bc2c91d65589ed456208011ba9a8ac255aa52332c2daea57a9d55215f75fb64", "generation_seconds_batch": 16.423975200974382, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5865700244903564, "target_z": 0.8671874247517409, "ce": 5.77639102935791, "genuineness": 4.770940780639648, "identity": 0.5749971270561218, "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/vn_STNC_reg_low/87ed21416792ca114796bc43f92d889b108f2ba2874f73eafd4c2040d327ec42.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "87ff0ff2e09593522fc58e409700565a58833a27102c783a5391d2274b9c082f", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "08de1293a81bab0bdfe20a6c491d2069e71b5e1df27eb2a5bcea8911485ddb36", "generation_seconds_batch": 21.688895749044605, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.356050491333008, "target_z": 0.19140623527751452, "ce": 6.300607204437256, "genuineness": 4.381826400756836, "identity": 0.3143694996833801, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/87ff0ff2e09593522fc58e409700565a58833a27102c783a5391d2274b9c082f.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "8854876b6d9089e1da072dcd5ec32b2b92efd376d67e8d64e591bf50a4cb0e42", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d0417104a1a916e21673dcc18fe84fb2a254d66f4823e2c431bf073699a95980", "generation_seconds_batch": 17.052191909984685, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7733802795410156, "target_z": 0.7031249672833656, "ce": 5.808161735534668, "genuineness": 4.419482707977295, "identity": 0.68414705991745, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/8854876b6d9089e1da072dcd5ec32b2b92efd376d67e8d64e591bf50a4cb0e42.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "892c992b437ad851768c7adcd8184e84dd59cfa6a2b1c55850ddeadaecb695f7", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "184ce5b3f505e324c22ca75ec0adaed0720e3607b27d58b4194ebffbd4496253", "generation_seconds_batch": 22.858420604025014, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6933186054229736, "target_z": 0.7734375687049322, "ce": 5.700935363769531, "genuineness": 4.256305694580078, "identity": 0.7704737186431885, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/892c992b437ad851768c7adcd8184e84dd59cfa6a2b1c55850ddeadaecb695f7.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "892eb9e73aba836582fc90a7885d73bd2445edab658f98c6a771d623e4fbb69b", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2f8e389e1dd8aa8efc48859f0bd8ea9e1e9ed7916aa2358d8814b0af1c80b1cf", "generation_seconds_batch": 21.11886638798751, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3730725049972534, "target_z": 1.0546874509250483, "ce": 5.6532793045043945, "genuineness": 4.607763767242432, "identity": 0.7267277240753174, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/892eb9e73aba836582fc90a7885d73bd2445edab658f98c6a771d623e4fbb69b.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "895238ab5f6c67c4d6d75c04a72f75713d95ce565f409d3bf2e2892eeaf688ff", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8bd08eacc1e57d342b6da4c433904b8b5bb3403a150a94502c5fd9cab110f602", "generation_seconds_batch": 16.030564768007025, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9690862894058228, "target_z": 0.5312499869133462, "ce": 5.764476776123047, "genuineness": 4.557555198669434, "identity": 0.7377487421035767, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. J'ai 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/vn_STNC_reg_low/895238ab5f6c67c4d6d75c04a72f75713d95ce565f409d3bf2e2892eeaf688ff.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "8993581e778bd6ac12fc6a21bf11a9b6c2501fb9a13d576cc5a0ccb4abfb1f6c", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "439966ae3f548824c700ca60e388afc7db47ecff1cd5ee92341d33a99a871253", "generation_seconds_batch": 16.675345901981927, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7185513973236084, "target_z": -0.1269531045521035, "ce": 6.3641486167907715, "genuineness": 4.6203155517578125, "identity": 0.5783629417419434, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/8993581e778bd6ac12fc6a21bf11a9b6c2501fb9a13d576cc5a0ccb4abfb1f6c.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "89f71724239d0b8b0e17b8863a7db9b523f9b15db43f1d1987af8dd958eebd62", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6cc8d5470d31fd6e51e49d41def8ab510076865c6559ce69248c6600cc3cc819", "generation_seconds_batch": 17.199097343022004, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6287600994110107, "target_z": -0.04809565108510345, "ce": 6.475345611572266, "genuineness": 4.68307638168335, "identity": 0.7377758622169495, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/89f71724239d0b8b0e17b8863a7db9b523f9b15db43f1d1987af8dd958eebd62.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "8a16178cd037a64d0099b878be3ff8b14485624439773d003db1ef9d83e9f499", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f571835a2566f17071f416838cbfcc648b6780a23d33659a2222d5718ad04843", "generation_seconds_batch": 21.65992965293117, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0350347757339478, "target_z": 1.3515624836416829, "ce": 5.895530700683594, "genuineness": 4.545003414154053, "identity": 0.48419496417045593, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/8a16178cd037a64d0099b878be3ff8b14485624439773d003db1ef9d83e9f499.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "8a27e790e107cd85c2f7ec708adee76c025b75cd263546fdbd0c4d45df40e9f6", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9e354fde0f81d6e7859cff59cb61fadef344c188989c8519d00686aad6d6e08a", "generation_seconds_batch": 22.437450910103507, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0973048210144043, "target_z": 1.2968750327166343, "ce": 5.927301406860352, "genuineness": 4.557555198669434, "identity": 0.7134569883346558, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/8a27e790e107cd85c2f7ec708adee76c025b75cd263546fdbd0c4d45df40e9f6.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "8a5fbc1186b2067962a53cfd1367c8bf79231d6360edab59a0c7f70618a608ee", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2617bbc0d11d0d3a17487fd41ff937ab0e03cddbb30952905ce67ea7e69d2b2c", "generation_seconds_batch": 22.48926167807076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1425528526306152, "target_z": 0.37890636614405204, "ce": 6.181467056274414, "genuineness": 4.557555198669434, "identity": 0.5879874229431152, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/8a5fbc1186b2067962a53cfd1367c8bf79231d6360edab59a0c7f70618a608ee.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "8a7cac9c5b10bd8f295a6de1cdf9a01125a3302e2e4a90dc95f9103e22ad863b", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4715d4bb4f795939528ba165ea42be340ecc08488dbdbff3fc74256c96bd30b2", "generation_seconds_batch": 17.39467202196829, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5509870052337646, "target_z": 0.8984375163583171, "ce": 5.998785495758057, "genuineness": 4.432034492492676, "identity": 0.7667595148086548, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/8a7cac9c5b10bd8f295a6de1cdf9a01125a3302e2e4a90dc95f9103e22ad863b.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "8ab5068e5b772156b58ceb696e9c65af3f5a37c203cc7ed86faa74e490566f26", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "54db5e72d65c5167c8876f859fd6e281d86c270985ed5c425e371987c68be7dd", "generation_seconds_batch": 22.81780348799657, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.559262275695801, "target_z": 0.012939455067550166, "ce": 6.23706579208374, "genuineness": 4.570107460021973, "identity": 0.7785540819168091, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/8ab5068e5b772156b58ceb696e9c65af3f5a37c203cc7ed86faa74e490566f26.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "8b726fcecd7a79d7e055985e427d10d6be6252bfc79c3212a96e72d8700f2a6f", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e3726297047e10167027f0bc6f345a7235a6f0ca04a8b2e5f82009c631a87a2e", "generation_seconds_batch": 23.311130142072216, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.585810422897339, "target_z": -0.010375936637982117, "ce": 6.205295085906982, "genuineness": 4.645420074462891, "identity": 0.5803937911987305, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/8b726fcecd7a79d7e055985e427d10d6be6252bfc79c3212a96e72d8700f2a6f.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "8b98fa0eb893a70056d6c8595875ff343fe47c293478ff38d962638d1337d3ae", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e2dea83f1bde98d7102b1dbdbc6a0a9bdad0456f8637555d66169082f01db33d", "generation_seconds_batch": 16.53627523209434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.079513430595398, "target_z": 1.3124999738266925, "ce": 6.268836498260498, "genuineness": 4.770940780639648, "identity": 0.3912104368209839, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/8b98fa0eb893a70056d6c8595875ff343fe47c293478ff38d962638d1337d3ae.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "8ba22862eeb7b7ee15f872274e4922bf340aebd6cbbcbe94a1a9735528611fe7", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ac05c6ff74a97e9f0c8b74304057f897b8209bb7141b1477a84daa2fae2a19a4", "generation_seconds_batch": 22.964052592986263, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7770586013793945, "target_z": 1.5781250196299805, "ce": 5.8875885009765625, "genuineness": 4.632867813110352, "identity": 0.49744319915771484, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/8ba22862eeb7b7ee15f872274e4922bf340aebd6cbbcbe94a1a9735528611fe7.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "8c51c76bc3eaff7518fa0185d4ce8cfb013c65347092cdeebf1df308ad3e68a1", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a38ccfd4de9c3c87dceed792b4bb7d1bef4df53caf3947ae538e96357af2ca0f", "generation_seconds_batch": 23.836005758028477, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.008347511291504, "target_z": 1.375000052346615, "ce": 5.220403671264648, "genuineness": 4.6203155517578125, "identity": 0.8550946712493896, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/8c51c76bc3eaff7518fa0185d4ce8cfb013c65347092cdeebf1df308ad3e68a1.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "8c79bb5814051f9f52a1e204b49c146f8b9e0cc227a34c4982d44a851a05ded5", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c9ec63514f3c11c3150cd5cafd2d5875b5fbcedcf442f02458f5ea4e8cb30023", "generation_seconds_batch": 21.898873863043264, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5687785148620605, "target_z": 0.8828124705550291, "ce": 5.895530700683594, "genuineness": 4.6705241203308105, "identity": 0.7090210318565369, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/8c79bb5814051f9f52a1e204b49c146f8b9e0cc227a34c4982d44a851a05ded5.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "8c9c4efa664778b15cc77969a16ef91e870be87689bd814a5e26e4167600133a", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8ad3a3ca3d18877dba979c55041abe3e943a7e0043a58b9ed61280c2844cb960", "generation_seconds_batch": 21.431503625004552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0958504676818848, "target_z": 0.41992182347130086, "ce": 6.23706579208374, "genuineness": 4.545003414154053, "identity": 0.579824686050415, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/8c9c4efa664778b15cc77969a16ef91e870be87689bd814a5e26e4167600133a.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "8ce57c8f9f3beea147ab77edeb8f25eb8f4b30db4c199371ed66b5cc13e5ce86", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "177d8d8d47d88535b7e8f768404ac0afa0bf42919c4e9d00a6464cdea6a83dab", "generation_seconds_batch": 22.877725863014348, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7689323425292969, "target_z": 0.7070312810808026, "ce": 5.935244083404541, "genuineness": 4.419482707977295, "identity": 0.7487305402755737, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/8ce57c8f9f3beea147ab77edeb8f25eb8f4b30db4c199371ed66b5cc13e5ce86.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "8cf353138f2bfefe1cfe652c68d8a3c3a4faa3baed524c201bea2580a74b3a5e", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ddfe6042a88d39395e3d35e36d89346ce1a00df078a2f990858219ae309cd88b", "generation_seconds_batch": 16.766530635999516, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.053595781326294, "target_z": 0.4570311763875727, "ce": 6.054384231567383, "genuineness": 4.65797233581543, "identity": 0.6396926641464233, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/8cf353138f2bfefe1cfe652c68d8a3c3a4faa3baed524c201bea2580a74b3a5e.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": "8d11ba56c7ec14c91da0c221c05df6e1f8a12310eebe00512d91036024f0cef9", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "43337eb7ba01428a35433b0d80b4fc09915e68850432c9794b7c4d0f267c1af1", "generation_seconds_batch": 20.724322946043685, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5154041051864624, "target_z": 0.9296875032716634, "ce": 5.804190158843994, "genuineness": 4.570107460021973, "identity": 0.7044028043746948, "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/vn_STNC_reg_low/8d11ba56c7ec14c91da0c221c05df6e1f8a12310eebe00512d91036024f0cef9.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "8d53223bbbe6b5a83215a50cd9fe3cb1a5a582d8dbce8e86005ef4533148abb1", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f65209d74ae0f2c578625fa04d52a39c4533ed36f31e926c205c2f0634a2ac48", "generation_seconds_batch": 22.434688107925467, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.639944314956665, "target_z": 0.8203124967283366, "ce": 6.133810997009277, "genuineness": 4.457139015197754, "identity": 0.4284706115722656, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/8d53223bbbe6b5a83215a50cd9fe3cb1a5a582d8dbce8e86005ef4533148abb1.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "8df6cedb7a6e8c1fe63a5407912b1136a1aa7a1b14c4ff6f5f7657c5d40d6ef6", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6a1176a8ef37635ca9f0e52a8f89f57642d0c7efcdcfaa608837c7ec1cd919cf", "generation_seconds_batch": 16.681532692047767, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8660159111022949, "target_z": 1.5, "ce": 6.0146708488464355, "genuineness": 4.482243061065674, "identity": 0.5692512392997742, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/8df6cedb7a6e8c1fe63a5407912b1136a1aa7a1b14c4ff6f5f7657c5d40d6ef6.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "8df9c9a05fc654a3043e02a4c270f02c29c17c248cb1758f43ea32c0a77782ff", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "55f2b2f55504cc6387aa999efa697d9f64b36b66ee65f89051ef485fa8437cea", "generation_seconds_batch": 23.05403789097909, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7066622972488403, "target_z": 0.761718732005851, "ce": 6.078212261199951, "genuineness": 4.595211505889893, "identity": 0.8341217041015625, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/8df9c9a05fc654a3043e02a4c270f02c29c17c248cb1758f43ea32c0a77782ff.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "8dff5cdee7e4ff340cc97f94e64561f8a3e5a06b73b3bc11a1e3fd98a518a25d", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "10f990010a2159772a7513eb21fd5b661807dab8392bbfeedd7d41b62b7c0623", "generation_seconds_batch": 17.65638886601664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.477810859680176, "target_z": 0.0844725701643558, "ce": 5.847874641418457, "genuineness": 4.231201171875, "identity": 0.6434852480888367, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/8dff5cdee7e4ff340cc97f94e64561f8a3e5a06b73b3bc11a1e3fd98a518a25d.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": "8e0ef16f4f6abc8bcd562f36bbb2a61a810f73c49e2f3927f014d01748ff7f77", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The road turns left behind the bridge.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ee6476a092d201d2bef1255b03c8f1a8b2c07829f3e5182e9ea0357f2fd80c6e", "generation_seconds_batch": 17.330610017990693, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8979203701019287, "target_z": 0.5937500654332687, "ce": 5.577824115753174, "genuineness": 4.256305694580078, "identity": 0.6622180342674255, "hypothesis": "The road turns left behind the bridge. Keep going until you r reach the small wooden gate. I will meet you on the other side.", "wer": 0.041666666666666664, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/8e0ef16f4f6abc8bcd562f36bbb2a61a810f73c49e2f3927f014d01748ff7f77.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "8e594eba07f07642652576852921beaafcf38ab8f04c811d39706df0b3547c90", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5e3e1d2985182bb91215ec3e32938cde59c481111c9dbfd0383fbad46d6692bd", "generation_seconds_batch": 22.53774720407091, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6703099012374878, "target_z": 1.6718749803700195, "ce": 5.720792293548584, "genuineness": 4.557555198669434, "identity": 0.5243746042251587, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/8e594eba07f07642652576852921beaafcf38ab8f04c811d39706df0b3547c90.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "8ea0eb81fd9ed9d14383edb0202b188d91ad0e9ec0f2f3d5536110d6ff6fdc96", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a6d558149939705b38182d0381d36453ce245950791901e056b9d0de4bba3bf4", "generation_seconds_batch": 16.169855204992928, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2396365404129028, "target_z": 1.1718749803700195, "ce": 5.9431867599487305, "genuineness": 4.921566009521484, "identity": 0.6948179006576538, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/8ea0eb81fd9ed9d14383edb0202b188d91ad0e9ec0f2f3d5536110d6ff6fdc96.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "8eb0ff1ab12aa71d5c00b11abedce20e7ebce5402b4f3e7c682cd3b9e2093546", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6c71d898d9c8c45eeb02e2982abcbc0fae2d55e433af50ed09d0eba2f132933a", "generation_seconds_batch": 17.27809655899182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.591231346130371, "target_z": -0.015136756578608492, "ce": 6.284721851348877, "genuineness": 4.482243061065674, "identity": 0.7377753257751465, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/8eb0ff1ab12aa71d5c00b11abedce20e7ebce5402b4f3e7c682cd3b9e2093546.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "8eb34ef3d3364e05a64eb96600d9e9bc8ad032980de06248a1be2e73df500bb5", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "62182d3f3e8114edd3c82fbe177e638e78ef2adcc78c8fdc9f418b17d32f0815", "generation_seconds_batch": 23.34184319106862, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.671079397201538, "target_z": 0.7929687189191974, "ce": 5.6413655281066895, "genuineness": 4.607763767242432, "identity": 0.4816710352897644, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/8eb34ef3d3364e05a64eb96600d9e9bc8ad032980de06248a1be2e73df500bb5.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "8f2d40e8d0e2aadac5104c9501692c5d825ab51f98607d22a0cf413fd62e6d57", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a10857d3c0f4b0d30ca354f10c19c4b5a025b3386c3aa05a74b81867eb95d9d2", "generation_seconds_batch": 23.224018677021377, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0973048210144043, "target_z": 1.2968750327166343, "ce": 6.046441555023193, "genuineness": 4.557555198669434, "identity": 0.4799045920372009, "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/vn_STNC_reg_low/8f2d40e8d0e2aadac5104c9501692c5d825ab51f98607d22a0cf413fd62e6d57.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "8f704d4da35694c9d07d909581594f8b97291fc3ad51cafbde40aead7da3b412", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "6c01e195e2209a1e848d4da6abeff6bf51828e2233df853aa8c1d4008871a592", "generation_seconds_batch": 22.336110053001903, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.221845030784607, "target_z": 1.1875000261733075, "ce": 5.871703147888184, "genuineness": 4.545003414154053, "identity": 0.49539971351623535, "hypothesis": "The 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/vn_STNC_reg_low/8f704d4da35694c9d07d909581594f8b97291fc3ad51cafbde40aead7da3b412.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "8fb2f9b31772666257a18880a79aa47c491eca02e8b8b3444375f8a58a76b867", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8112cf02f6d9a44745c2ecf75be4cfd2b78885ab4dc5952a0f6acc634fcd2179", "generation_seconds_batch": 22.832411274896003, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3552809953689575, "target_z": 1.0703124967283366, "ce": 5.538110733032227, "genuineness": 4.519899368286133, "identity": 0.7161417007446289, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/8fb2f9b31772666257a18880a79aa47c491eca02e8b8b3444375f8a58a76b867.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "90314c0af0056dc4c6cf3adc85adfe87f3d4be7689ce89b520705021f09c8d36", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "65b5718f03e2d76a5047253e8a2168fcf128b887cfce9336f6380a291ef81edc", "generation_seconds_batch": 21.496552930912003, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7111101150512695, "target_z": 0.757812522901644, "ce": 5.967014789581299, "genuineness": 4.362998008728027, "identity": 0.41333287954330444, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/90314c0af0056dc4c6cf3adc85adfe87f3d4be7689ce89b520705021f09c8d36.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "90c4446d47246defc546a328fef0b9fd45df8dd0fbdaa5fa692d0a4c88595a21", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7060ba8d4b45e34d57a8c9319ac28ef07f7a7b175b4c7c69c6abdb4e499f56c4", "generation_seconds_batch": 17.34895976597909, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2381820678710938, "target_z": 0.29492187581791585, "ce": 6.0305562019348145, "genuineness": 4.645420074462891, "identity": 0.6024733781814575, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/90c4446d47246defc546a328fef0b9fd45df8dd0fbdaa5fa692d0a4c88595a21.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "915b27c534ac5d06baaa5d93bbd08e8bb3783f1535444e193134883f17c36057", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "da0915be1dda8fcec72faff61f3d041dd306a644ffc08aac484e818181a888cb", "generation_seconds_batch": 21.631899173022248, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8675549030303955, "target_z": -0.2578125229016441, "ce": 6.53094482421875, "genuineness": 4.7081804275512695, "identity": 0.714972972869873, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/915b27c534ac5d06baaa5d93bbd08e8bb3783f1535444e193134883f17c36057.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "91f9588eef591723520e46343207bc8fe7dbd3dbfffd7ef32ffcffadb27cb7ac", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fffa805a0cc5097b170c33d6b35275896fd16bdb9013063f34adba7991a18d72", "generation_seconds_batch": 22.393209049012512, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2737650871276855, "target_z": 0.26367178421133963, "ce": 5.871703147888184, "genuineness": 4.3504462242126465, "identity": 0.7131825685501099, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/91f9588eef591723520e46343207bc8fe7dbd3dbfffd7ef32ffcffadb27cb7ac.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "9263fb5c89a5193bcf950dd7323096fe73df5660c192559181942bd6c62c6e0e", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1e031f80c1e43e1a5ceb99424e5148181b88bad5c4ac667dd175d92a183ed4c8", "generation_seconds_batch": 17.022994811995886, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8838074207305908, "target_z": 1.484374954196712, "ce": 5.8875885009765625, "genuineness": 4.6956281661987305, "identity": 0.039301857352256775, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/9263fb5c89a5193bcf950dd7323096fe73df5660c192559181942bd6c62c6e0e.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "92f1d7203369f53a2b2ac43464328d6c5dbe01b099d94b936c54346e9189f8fb", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e5706df1c0febb216772d601de1f7336d37ccb83176715f87097caa9dd81c26d", "generation_seconds_batch": 22.718832734040916, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8126415014266968, "target_z": 1.5468750327166343, "ce": 6.038498878479004, "genuineness": 4.607763767242432, "identity": 0.650652289390564, "hypothesis": "Ik 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/vn_STNC_reg_low/92f1d7203369f53a2b2ac43464328d6c5dbe01b099d94b936c54346e9189f8fb.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "93207bd6964726c856c813d9a88b0b14bcaf18b870d13be24930d3c48503f3fa", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ca12e7f15bcc541ff202ebb25ac61fbc85a5595a2b1f36440deee2c9ce9ebe4a", "generation_seconds_batch": 16.524563557002693, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.942399024963379, "target_z": 0.5546875556182784, "ce": 6.022613525390625, "genuineness": 4.846253395080566, "identity": 0.7397212386131287, "hypothesis": "He encontrado la cajita, estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/93207bd6964726c856c813d9a88b0b14bcaf18b870d13be24930d3c48503f3fa.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "9327db57da957d3f35a6ae3715efd010d69540c5c978b706df280a2ac692159f", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4cbffbe83def400f74910e9e6437f98fa18ee9ec8a3a5f8c1656a05c76c1fa5b", "generation_seconds_batch": 17.37296420300845, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1595749855041504, "target_z": 1.2421874770983559, "ce": 5.637393951416016, "genuineness": 4.846253395080566, "identity": 0.7666906118392944, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/9327db57da957d3f35a6ae3715efd010d69540c5c978b706df280a2ac692159f.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "938074edf76b4b3a7d08559b27f5fbc7bc468ea6073bbfb90acd7741bde9132f", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2ad89dc60a434aa0f60c4fc00de23684a07a22de47e4f1f218d00a69e5532ef7", "generation_seconds_batch": 22.220634363009594, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.004669189453125, "target_z": 0.5, "ce": 5.935244083404541, "genuineness": 4.118232727050781, "identity": 0.525211751461029, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/938074edf76b4b3a7d08559b27f5fbc7bc468ea6073bbfb90acd7741bde9132f.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "93bb57d3fb5c8d774dacf17dcd5150eb34cd1aa1873168e5e726fdb986d86b0f", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "117c4ddae64556a35b8a4cc21e036146d7ce24666bfe6c603be3f4a6f701b1e3", "generation_seconds_batch": 25.939192579942755, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.386073589324951, "target_z": 0.16503903591773458, "ce": 6.0305562019348145, "genuineness": 4.582659721374512, "identity": 0.74882972240448, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/93bb57d3fb5c8d774dacf17dcd5150eb34cd1aa1873168e5e726fdb986d86b0f.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": "941ca9767d374adc50484cc18dd89373a2dbe93b2eaf9d68548494e1b5e24e22", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "19cc4d2de181f4d6917cc80e965f743f972096621542796927f3e9cbde3d269a", "generation_seconds_batch": 20.893695896025747, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5287476778030396, "target_z": 0.9179687712658123, "ce": 5.422942161560059, "genuineness": 4.645420074462891, "identity": 0.7396829128265381, "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/vn_STNC_reg_low/941ca9767d374adc50484cc18dd89373a2dbe93b2eaf9d68548494e1b5e24e22.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "94605fadf0ae39c2e6e2c83e29a2e4444ca2494a6cf95ca34717b0d417a9aaf6", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "86dbac9336bba1088bc1c11833e10f8da13b62078269b3fa548be22a3bdb14dc", "generation_seconds_batch": 22.49646178993862, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0343499183654785, "target_z": -0.4042969870544727, "ce": 6.41180419921875, "genuineness": 4.337893962860107, "identity": 0.5820420384407043, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/94605fadf0ae39c2e6e2c83e29a2e4444ca2494a6cf95ca34717b0d417a9aaf6.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "94e4a4f2f409407b161a5cca6714c44ec4bad4c1610a533e28ac44cd34984ddd", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b2a28a1f97a202d0936b4bfc04ac23c027d876677b6b155941e3c0ea717c757e", "generation_seconds_batch": 22.220634363009594, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5598827600479126, "target_z": 0.8906249934566731, "ce": 6.078212261199951, "genuineness": 4.457139015197754, "identity": 0.4309689998626709, "hypothesis": "The 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/vn_STNC_reg_low/94e4a4f2f409407b161a5cca6714c44ec4bad4c1610a533e28ac44cd34984ddd.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "950566716f3e622ce53ad642f1e9fa8595323a8cdd4e38549ed286fe3ba000fd", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2591e2895e515fd44a4b047676f407e7237368f844f129693213f7e73ce1c589", "generation_seconds_batch": 21.65992965293117, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3285938501358032, "target_z": 1.0937499607400387, "ce": 5.879645347595215, "genuineness": 4.582659721374512, "identity": 0.6412036418914795, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/950566716f3e622ce53ad642f1e9fa8595323a8cdd4e38549ed286fe3ba000fd.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "9516cc01897a7265932034a01780696a834cf1c007553b923361bc151556866e", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "3e0670f679863d88c0583f04d97b2ace0090e1430976444c82d7e25ddfcad99c", "generation_seconds_batch": 22.24303964106366, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.43534255027771, "target_z": 1.0, "ce": 5.764476776123047, "genuineness": 4.720732688903809, "identity": 0.6819645762443542, "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/vn_STNC_reg_low/9516cc01897a7265932034a01780696a834cf1c007553b923361bc151556866e.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "956d5329007d3a247583bfa532c28603ca77af266045aa65829e1c86f039ba13", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "580ad7c083d04a40d283767716bb3a2280a60ea27b776d89a87eb5efebfec0ae", "generation_seconds_batch": 23.433784519904293, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.104746103286743, "target_z": 0.4121094052628868, "ce": 6.0146708488464355, "genuineness": 4.507347106933594, "identity": 0.6795467138290405, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/956d5329007d3a247583bfa532c28603ca77af266045aa65829e1c86f039ba13.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "95aab50c0a5201701fff645e31e63401f2f5e73c3e5103c4ebcb053e2d8c6455", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "45c7390fbc629524ca6b227a7b61f964615922edea5ddaf944f5cdddc71bbb9b", "generation_seconds_batch": 23.392996742972173, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.221845030784607, "target_z": 1.1875000261733075, "ce": 5.847874641418457, "genuineness": 4.394378662109375, "identity": 0.17652516067028046, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/95aab50c0a5201701fff645e31e63401f2f5e73c3e5103c4ebcb053e2d8c6455.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "96397f388d9cd2baecf4c8196c205393fe15c161f7bd88ffd613ed42fb2eff3c", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "db9ab8c4cc94707e286656c5cbca6ffbec9575d5d036c15386f7ad6de860910a", "generation_seconds_batch": 21.809175229049288, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0246846675872803, "target_z": 0.4824217972979934, "ce": 5.959072113037109, "genuineness": 4.482243061065674, "identity": 0.6251680850982666, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/96397f388d9cd2baecf4c8196c205393fe15c161f7bd88ffd613ed42fb2eff3c.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "97ebac32a9ed2c9a35272f0fa9a2b9e9ad632c6dfe00af63d3818ec45c3d38fb", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c3559763f5ca6a283ffd199b73c5c2fb85808a088a43b0b20541a04df69de19a", "generation_seconds_batch": 17.706129434052855, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.744126558303833, "target_z": -0.14941399011444645, "ce": 6.1020402908325195, "genuineness": 4.457139015197754, "identity": 0.6521837711334229, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/97ebac32a9ed2c9a35272f0fa9a2b9e9ad632c6dfe00af63d3818ec45c3d38fb.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": "98041b913d4e2b70fd6cd01955d0506a1c6b58e93d4d9859f6b328aaab008d91", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9bfaa241d6f149453588298f1fca1896eebef79d0c528de0d09811767dd564f1", "generation_seconds_batch": 17.543925322010182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0261390209197998, "target_z": 1.3593750065433268, "ce": 5.585766792297363, "genuineness": 4.545003414154053, "identity": 0.8720026612281799, "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/vn_STNC_reg_low/98041b913d4e2b70fd6cd01955d0506a1c6b58e93d4d9859f6b328aaab008d91.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "981646dca858775fff3e753d4bfec1457fc18d531b3272a6cdeeef0f49760527", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a4f45bec8d55367a1085df7be6dabf17c34120a0b2da6e50569401dc6d2fa062", "generation_seconds_batch": 16.681532692047767, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0973048210144043, "target_z": 1.2968750327166343, "ce": 5.871703147888184, "genuineness": 4.582659721374512, "identity": 0.39073118567466736, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/981646dca858775fff3e753d4bfec1457fc18d531b3272a6cdeeef0f49760527.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": "984aa9eb9a65b92601641ea32ea09dd5808e8f3a1ee016b68a836c430c1057d3", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c6c47d4e87c90ba26c67c22203b22f16e197479c4e1af9539f8c005cdd579291", "generation_seconds_batch": 20.5348345200764, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0439305305480957, "target_z": 1.3437499607400387, "ce": 5.696964263916016, "genuineness": 4.7081804275512695, "identity": 0.6074539422988892, "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/vn_STNC_reg_low/984aa9eb9a65b92601641ea32ea09dd5808e8f3a1ee016b68a836c430c1057d3.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "989c0a1162ac7ed9b370291f8bfc25a9f2e76a53d523b5784e348ffe865077ea", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "817166f30043c8d0d577b332cd048b1245546fef7a0aadd531000f7b3cb90790", "generation_seconds_batch": 21.631899173022248, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0232300758361816, "target_z": -0.39453120256088015, "ce": 6.332377910614014, "genuineness": 4.457139015197754, "identity": 0.5349079370498657, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/989c0a1162ac7ed9b370291f8bfc25a9f2e76a53d523b5784e348ffe865077ea.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": "9915f102bc172f3caff8ed9ec8e26454b85b4216fa6214f4c53bf0c2d29fba41", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "06a351b1ac11814235737b109a123d7546592396ec3d0b8f7dc9adf7232a8e37", "generation_seconds_batch": 20.415171385044232, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2396365404129028, "target_z": 1.1718749803700195, "ce": 5.7684478759765625, "genuineness": 4.6705241203308105, "identity": 0.24781464040279388, "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/vn_STNC_reg_low/9915f102bc172f3caff8ed9ec8e26454b85b4216fa6214f4c53bf0c2d29fba41.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "8bb6586f8948d12aa9c3886945eb8794ecafaad4d8cd679372905d3ad27094b8", "generation_seconds_batch": 16.690406416077167, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.698535919189453, "target_z": -0.10937490185009688, "ce": 6.300607204437256, "genuineness": 4.582659721374512, "identity": 0.656273603439331, "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/vn_STNC_reg_low/99bf1b7d3eed7b885fdae888ab00b4960791c3971d259b46f771306d2801e39b.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "9aa9d20df0c3fe9a07fb3b230fe3f3ae0e0fa718327ae2dde8f433cdc31ae5bd", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f52b9da75980abd26328c1f0391a048c69acb089113e6cd4411cd16852a25cdd", "generation_seconds_batch": 23.948627518955618, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2537496089935303, "target_z": 0.28124998691334624, "ce": 6.173524379730225, "genuineness": 4.632867813110352, "identity": 0.5651407837867737, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/9aa9d20df0c3fe9a07fb3b230fe3f3ae0e0fa718327ae2dde8f433cdc31ae5bd.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "9b1c4a80e80d2739d7597b76fc9d683a643f396ca50b262d29818dcda97fc892", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "5d0571c8a84e51e8b7202b1e1af17a65aa83f9f791ebd70f35c12424e7842bfa", "generation_seconds_batch": 22.056031152023934, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2292864322662354, "target_z": 0.30273429402632995, "ce": 6.197352409362793, "genuineness": 4.406930446624756, "identity": 0.36364102363586426, "hypothesis": "The 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/vn_STNC_reg_low/9b1c4a80e80d2739d7597b76fc9d683a643f396ca50b262d29818dcda97fc892.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "9b4eaa0e60d2d14dff8695f0861ab8c9b4df4a0dd3047ac81809775f90385a0c", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "112659b509be602063eaddb6283b9b50c80e49c078441311c5f378a4973fb5b6", "generation_seconds_batch": 17.201869227108546, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2537496089935303, "target_z": 0.28124998691334624, "ce": 6.09409761428833, "genuineness": 4.419482707977295, "identity": 0.5348349809646606, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/9b4eaa0e60d2d14dff8695f0861ab8c9b4df4a0dd3047ac81809775f90385a0c.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "9bd230356a1be27a22ebc219b4030b9eda9d53d8ad4d53174ca5e6ba141e6803", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "db94a9d42152593bc4b33f3947bbe284d61ec685c515a3118733f9d66749aa39", "generation_seconds_batch": 16.20014598907437, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.092172145843506, "target_z": -0.45507822887531413, "ce": 6.157639026641846, "genuineness": 4.570107460021973, "identity": 0.6038807034492493, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/9bd230356a1be27a22ebc219b4030b9eda9d53d8ad4d53174ca5e6ba141e6803.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "9bf20352845bd891c5d7034666eed4b9047b31211b056d0062d7da2deb8cc0f0", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ac68f5eabc98e5e63bcf81c1808f1bf2d11c39f584b10616cd54e7ffd90710dc", "generation_seconds_batch": 22.605406500049867, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1981513500213623, "target_z": 0.33007807183546917, "ce": 6.078212261199951, "genuineness": 4.507347106933594, "identity": 0.40828222036361694, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/9bf20352845bd891c5d7034666eed4b9047b31211b056d0062d7da2deb8cc0f0.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "9c07997a2e97d7ef5cba4da6f47f934a728e8b01ad1a17d62c6d8399fc768ebc", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b3a5bdba4af9cb4e73c38411c8f73e3cb865d1d718f3fc459bf81d8618c66e8b", "generation_seconds_batch": 16.535791578004137, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9468469619750977, "target_z": 0.5507812418208414, "ce": 6.0146708488464355, "genuineness": 4.68307638168335, "identity": 0.7284988760948181, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/9c07997a2e97d7ef5cba4da6f47f934a728e8b01ad1a17d62c6d8399fc768ebc.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "884bb2fd48a8c8a3a1e3c6fc4d88227c27b34a6f97c634d6cf97d38dee197703", "generation_seconds_batch": 17.883747690008022, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.236727714538574, "target_z": -0.5820313334274176, "ce": 6.181467056274414, "genuineness": 4.256305694580078, "identity": 0.6639620065689087, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de ooge boekenkast. We zetten het na de lunch terug.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/9c09e940c5dab53287ffae2d1a8fa5296f41b2df71714656aaa2bb8830bf5162.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "9c498e5d91ca88b30a54a33dd2bcf2019ab4e0912bd955f89343d83ed1835826", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fcc6f5a3c21256b180042d983f8b1423aa82dd8263edf9d652cad517cd2e3588", "generation_seconds_batch": 23.60449260403402, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9512948989868164, "target_z": 0.5468749280234044, "ce": 6.252951145172119, "genuineness": 4.65797233581543, "identity": 0.5733290314674377, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/9c498e5d91ca88b30a54a33dd2bcf2019ab4e0912bd955f89343d83ed1835826.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "9c5792793507f60dea512fa3482ebcc2476413ff6ea22603034c9e15591d0278", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4a253522a923d0573ff474edf9cf081d09f4c3d7f1385238d1924c6fb988b5b3", "generation_seconds_batch": 16.15646598499734, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9779820442199707, "target_z": 0.5234374640117022, "ce": 6.157639026641846, "genuineness": 4.758388519287109, "identity": 0.716127872467041, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/9c5792793507f60dea512fa3482ebcc2476413ff6ea22603034c9e15591d0278.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "7b6206fa91bb2631ce6b80cead33917de461187b67bba3dfeeadacfca0f23815", "generation_seconds_batch": 17.956805411027744, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0543651580810547, "target_z": -0.42187498037001936, "ce": 6.062326908111572, "genuineness": 4.394378662109375, "identity": 0.643118143081665, "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/vn_STNC_reg_low/9c969db20a6db380e08592f6aed9c9189b3db634d677ebe8f573f593ebdfd2d3.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "9cae6a2a40c8d727baf2f0e456b0748ae7473f6e6e788370ee0ba2536350ca9c", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f96ba1665c2994f499fde2e211a0400d0bb0bcfe5f1bc64487c43727fb0145f0", "generation_seconds_batch": 22.377440257929265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7708139419555664, "target_z": -0.1728516635126086, "ce": 6.197352409362793, "genuineness": 4.457139015197754, "identity": 0.5685643553733826, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/9cae6a2a40c8d727baf2f0e456b0748ae7473f6e6e788370ee0ba2536350ca9c.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "9cce002967fcc7c8b907e6875388dabc70b64cd0bb43fc8cf82be03b1b9baf8d", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bad201b1ee270c0c61c693a21124e371c781394a3d79cdade1876f551e6bafd1", "generation_seconds_batch": 16.38734126102645, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4976125955581665, "target_z": 0.9453125490749515, "ce": 5.760505676269531, "genuineness": 4.507347106933594, "identity": 0.6644489169120789, "hypothesis": "Ik heb het kleine doosje gevonden. Stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/9cce002967fcc7c8b907e6875388dabc70b64cd0bb43fc8cf82be03b1b9baf8d.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "9cd1a5ae602c9309378d771f17fd368909151a10db2863fc860204f8db750f06", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0cafd2fc4de7629063ed39b2c471e3bfefb47ebbdc55d1aadb965ce86abc96af", "generation_seconds_batch": 23.679422381101176, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2670931816101074, "target_z": 0.26953125490749513, "ce": 6.09409761428833, "genuineness": 4.419482707977295, "identity": 0.3509184718132019, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/9cd1a5ae602c9309378d771f17fd368909151a10db2863fc860204f8db750f06.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "9cf348b220b6991e30fdb78a5e96f6bc548d775db7811cd8c46fa97c7291ded5", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d2409f318c208a5d4a475314ffc4945ffe4cb08fefaf2c21734e13f680c39878", "generation_seconds_batch": 22.147765316069126, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3819681406021118, "target_z": 1.0468750327166343, "ce": 5.418970584869385, "genuineness": 4.168440818786621, "identity": 0.5747864246368408, "hypothesis": "The 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/vn_STNC_reg_low/9cf348b220b6991e30fdb78a5e96f6bc548d775db7811cd8c46fa97c7291ded5.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "9d002344887a4a2659a46ae4860d385fc21bfef67c81eb3701d36013e3207a69", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ab83c29d4f7fb9e5e379a3debcb45449b303f2be9cf5cfaca9d6a4c822e019ad", "generation_seconds_batch": 22.49646178993862, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1833531856536865, "target_z": -0.5351561960175533, "ce": 6.284721851348877, "genuineness": 4.369274139404297, "identity": 0.6579262614250183, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/9d002344887a4a2659a46ae4860d385fc21bfef67c81eb3701d36013e3207a69.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "9d2ebafab7a9355ef582b257bc30c5a5314c5083d9cf4e3f8510cf34023dcafe", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9106ec65a1590675de571153d2349d5e8d29dd3652f687184ba8f30af04ad413", "generation_seconds_batch": 16.768817565985955, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.896466016769409, "target_z": -0.28320314381206474, "ce": 6.149696350097656, "genuineness": 4.406930446624756, "identity": 0.2038576602935791, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/9d2ebafab7a9355ef582b257bc30c5a5314c5083d9cf4e3f8510cf34023dcafe.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "9d48e6d3bb837ff5c1a0e547529eff04301d78c5c9c6205aad5436ca709e1c89", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "198b6a118e4309bbd68c46d2ebc31ebdadc61b738c65b52162f6370582fb2b1f", "generation_seconds_batch": 22.30166957806796, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3552809953689575, "target_z": 1.0703124967283366, "ce": 5.879645347595215, "genuineness": 4.7458367347717285, "identity": 0.7025330662727356, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/9d48e6d3bb837ff5c1a0e547529eff04301d78c5c9c6205aad5436ca709e1c89.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "9d862de2089225878909c34e4876bc15de39000a43e5812e536f79d662718b02", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d4aa9aa0f82c3d1ce251b78c9f7339a5f289ec2be25efbd7e44df9f63b86bc5a", "generation_seconds_batch": 23.410492461989634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8267545700073242, "target_z": 0.6562500392599613, "ce": 6.0146708488464355, "genuineness": 4.457139015197754, "identity": 0.7422069311141968, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/9d862de2089225878909c34e4876bc15de39000a43e5812e536f79d662718b02.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "9d8fffc3055fcc56ffdd4d0df8ab97419a3b3557dfadc3fcee52ebdd3ac5addb", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1a7d6748aabb33c936e02a07c98f63f9b01f62195604e31d7bfde554fd9fa04c", "generation_seconds_batch": 22.30166957806796, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5420912504196167, "target_z": 0.9062500392599613, "ce": 5.720792293548584, "genuineness": 4.65797233581543, "identity": 0.7187790870666504, "hypothesis": "He encontrado la cajita. estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/9d8fffc3055fcc56ffdd4d0df8ab97419a3b3557dfadc3fcee52ebdd3ac5addb.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "9daffd1ea8ffeaf5ba4980dd886120dbcf88d31818898c7486f0dfa7129b211f", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f4bcb85d45bca8052128896b5ffbfbb4cec030d470347aa769b358b8c545ba66", "generation_seconds_batch": 17.302102377056144, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2226145267486572, "target_z": 0.30859376472248545, "ce": 6.165581703186035, "genuineness": 4.6203155517578125, "identity": 0.2889961004257202, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/9daffd1ea8ffeaf5ba4980dd886120dbcf88d31818898c7486f0dfa7129b211f.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "9dccdad88d38bdda76d29ade9cde1c700b17401f615ab3d9a6ba16db02e4b0c9", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d9f51e367dc4c3e95f69418766203b73cb2e347d93f28e26bf16413f18e16321", "generation_seconds_batch": 17.343595108017325, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5465391874313354, "target_z": 0.9023437254625242, "ce": 5.268059730529785, "genuineness": 4.595211505889893, "identity": 0.6705735325813293, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/9dccdad88d38bdda76d29ade9cde1c700b17401f615ab3d9a6ba16db02e4b0c9.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "9dd442ef62d0afc94ac282ceb70f11b345d748d05cef564eb6a0f787b6d09aee", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5d6bec7028114a078c31d16b12b9c328c6ecb23a1960995e67f3eca628b030c4", "generation_seconds_batch": 17.135985934059136, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5865700244903564, "target_z": 0.8671874247517409, "ce": 6.038498878479004, "genuineness": 4.557555198669434, "identity": 0.6752374768257141, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/9dd442ef62d0afc94ac282ceb70f11b345d748d05cef564eb6a0f787b6d09aee.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "9e6dad446eb41c2cf8400c1f2475a8ab72f1a3a4ee6c5615b9b5ca6c4991edea", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6af5a7ceea67d43af5aed6f4be9d29961c6d50224e2eaa1d5543b2b548ac5887", "generation_seconds_batch": 21.696319633047096, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1239920854568481, "target_z": 1.273437464011702, "ce": 5.593709468841553, "genuineness": 4.632867813110352, "identity": 0.6314181089401245, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/9e6dad446eb41c2cf8400c1f2475a8ab72f1a3a4ee6c5615b9b5ca6c4991edea.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "9ea7315ec216132654b0951a0fe63828ddbe54b2167136cbc15d069246a68ee2", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4f1acee6121a5cbb5e01394d7f5521034d8ec8c94a967cb7aa440739b2ef0d22", "generation_seconds_batch": 23.826806705095805, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9015988111495972, "target_z": 1.4687500130866538, "ce": 5.375286102294922, "genuineness": 4.770940780639648, "identity": 0.7668848633766174, "hypothesis": "La porte est encore ouverte J'ai posé ton manteau sur la chaise dis-moi quand tu es prêt", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/9ea7315ec216132654b0951a0fe63828ddbe54b2167136cbc15d069246a68ee2.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "9eec34c0a70bd877d8387e80421708d14f37ce6dca86182e2dbd86bc724fca64", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "bd9e87c2ffba57c34cf5fa7049f7fafa0edb6831f484a8eb6ec60cc351bdca4e", "generation_seconds_batch": 17.013290524017066, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5598827600479126, "target_z": 0.8906249934566731, "ce": 5.855817794799805, "genuineness": 4.65797233581543, "identity": 0.658394455909729, "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/vn_STNC_reg_low/9eec34c0a70bd877d8387e80421708d14f37ce6dca86182e2dbd86bc724fca64.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "9eecae144b90b27c1bcbc5d265748d624ed63f15c410c99112c05cf0dfb2891e", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c4e27c956071a5e531ef7aa77f72e45bd08efbc32d11b1eea2b13f5badad7b1e", "generation_seconds_batch": 22.530962995020673, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.417551040649414, "target_z": 1.015625045803288, "ce": 5.538110733032227, "genuineness": 4.570107460021973, "identity": 0.757603645324707, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/9eecae144b90b27c1bcbc5d265748d624ed63f15c410c99112c05cf0dfb2891e.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "9ef2206d639c1600de5d35e6e861bf6d27f40663edb8f2b63fb011ffa4b35965", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5f32ad20910940638774ad569cc38af8ebb44d9bcc9eabd011e362a581946ef8", "generation_seconds_batch": 22.420944545068778, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5865700244903564, "target_z": 0.8671874247517409, "ce": 5.879645347595215, "genuineness": 4.432034492492676, "identity": 0.5235104560852051, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/9ef2206d639c1600de5d35e6e861bf6d27f40663edb8f2b63fb011ffa4b35965.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "9f4898e81c90db5053980495363676697ca422fe49966a087179c5b21f21405d", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fed1318b890ac531692fe52eaa6c96f9fbaf9775622d8daf45d4d3df88317637", "generation_seconds_batch": 22.206867497996427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.816404342651367, "target_z": -0.2128905423904982, "ce": 5.704906940460205, "genuineness": 4.231201171875, "identity": 0.6057028770446777, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/9f4898e81c90db5053980495363676697ca422fe49966a087179c5b21f21405d.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "9fd1772d4e531d2728951c5c627b7c85b50a5326af58cf91c955306bc77cb7d8", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "25e308ced2def9c89f40c000608c52b22f12e836b204df859475c18c2ce9ffe5", "generation_seconds_batch": 23.142148748971522, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0973048210144043, "target_z": 1.2968750327166343, "ce": 5.80021858215332, "genuineness": 4.921566009521484, "identity": 0.6244556307792664, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt ?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/9fd1772d4e531d2728951c5c627b7c85b50a5326af58cf91c955306bc77cb7d8.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "a006f39a4d59feedb8ad4287ab2bcca8f184a7420e691c4d9e5eeb206c90c150", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8809e8e44288a4585720a4c9049dde08eedf3dbf5a5c9c521a053de5052967da", "generation_seconds_batch": 22.719563339021988, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8534417152404785, "target_z": 0.6328125752482591, "ce": 6.197352409362793, "genuineness": 4.494794845581055, "identity": 0.5055100917816162, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/a006f39a4d59feedb8ad4287ab2bcca8f184a7420e691c4d9e5eeb206c90c150.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "a01a29f2cbacd3ecd4a0b1aeccfdf9913d5218036f2e78c920924297e14d9d66", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0eb5997a41d0624a3ff3c2383b0c2f61416bf39c0f9f7d89ff33a04c124615cf", "generation_seconds_batch": 22.7956814939389, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0261390209197998, "target_z": 1.3593750065433268, "ce": 4.954324245452881, "genuineness": 4.595211505889893, "identity": 0.6863083839416504, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/a01a29f2cbacd3ecd4a0b1aeccfdf9913d5218036f2e78c920924297e14d9d66.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "a0c3b6b2f6a444a73f82e931e8efb834f4647be1935f5951950746d1e56646d8", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fc9c8ae46c18f04660f1c89314d6fcadd78329e07aaa540c8713b9f54d1249a7", "generation_seconds_batch": 21.65205686201807, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3641767501831055, "target_z": 1.0624999738266925, "ce": 6.09409761428833, "genuineness": 4.632867813110352, "identity": 0.4241887032985687, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/a0c3b6b2f6a444a73f82e931e8efb834f4647be1935f5951950746d1e56646d8.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": "a15f4cec08691e472cb6aca1dc7e20f0619a6305a00b0f9871ee1e0b3040adaa", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Your message arrived after I had left.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c18204c858219efcbab4f9312cff8a6051752a282a03af4f191ea02e47b09325", "generation_seconds_batch": 22.753210394992493, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9646384716033936, "target_z": 0.5351561960175533, "ce": 5.927301406860352, "genuineness": 4.143336772918701, "identity": 0.5912538170814514, "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/vn_STNC_reg_low/a15f4cec08691e472cb6aca1dc7e20f0619a6305a00b0f9871ee1e0b3040adaa.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "a16c92c7278e4ce64de1afb0e636a2d8f63a7aadd5716103b39876142e6211a8", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3d20d75a1ec84f6efa1fd3b6342961b287511e01e682164deafa36a782e4dcd8", "generation_seconds_batch": 20.960406412021257, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2040536403656006, "target_z": 1.2031249672833657, "ce": 5.9431867599487305, "genuineness": 4.833701133728027, "identity": 0.7022748589515686, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/a16c92c7278e4ce64de1afb0e636a2d8f63a7aadd5716103b39876142e6211a8.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "a19302322320962dc9a5ad0fd6bc83bd3627f3213ce4cda6ff17f299b9bd44fd", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1c0c58c6ee0b81334f7a22a8154dce0a61fc64186ca1809fcaa6f779b059044d", "generation_seconds_batch": 20.126902670017444, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6596171855926514, "target_z": -0.07519528428190285, "ce": 6.133810997009277, "genuineness": 4.432034492492676, "identity": 0.6420516967773438, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás. ¿Listo?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/a19302322320962dc9a5ad0fd6bc83bd3627f3213ce4cda6ff17f299b9bd44fd.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "a1956f6315f10875790bf25cededdc949ab5e0be0afd14776150cc9f4ef444b1", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d33e924877a7bd4b276bca0606959bfe47a62af965b6b40aded715dbc74c55d6", "generation_seconds_batch": 22.236689200042747, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8089631795883179, "target_z": 0.6718749803700194, "ce": 6.0305562019348145, "genuineness": 4.394378662109375, "identity": 0.6693612933158875, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude oude raam. We kunnen er samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/a1956f6315f10875790bf25cededdc949ab5e0be0afd14776150cc9f4ef444b1.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "a19a4433c18d1f6a675b2054906daf46711b14786e5470eea5d719003bc686dd", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b4b74b9609a6b17c582d95a860ef81c5ec2271bb6a3a25db39dba79dcac831ee", "generation_seconds_batch": 23.43883040198125, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.639944314956665, "target_z": 0.8203124967283366, "ce": 5.871703147888184, "genuineness": 4.758388519287109, "identity": 0.5948094129562378, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/a19a4433c18d1f6a675b2054906daf46711b14786e5470eea5d719003bc686dd.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": "a1bdad5c5c1385a36636a362f55bc2295e20a59d9c8f00c7863a32a4f85f0d00", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f290d6c1840ade80514b39266a8816bc1e3ebe342949387586d2298fe0e79080", "generation_seconds_batch": 20.65967797394842, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1328877210617065, "target_z": 1.265625045803288, "ce": 5.395142555236816, "genuineness": 4.595211505889893, "identity": 0.7648391723632812, "hypothesis": "J'ai écrit le numéro sur l'enveloppe, les le bien et verified les two derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.13636363636363635, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/a1bdad5c5c1385a36636a362f55bc2295e20a59d9c8f00c7863a32a4f85f0d00.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "a1be9d157ea4b8105134d9135df8b567990b099a3040df8241b1fb4e8ad9c4e4", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "85e61a5f00c6e6db4dd48496e68abf8308f2a95e964b48d4a1160992802d5b1b", "generation_seconds_batch": 23.199607295915484, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5331956148147583, "target_z": 0.9140624574683753, "ce": 5.879645347595215, "genuineness": 4.6203155517578125, "identity": 0.688325047492981, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/a1be9d157ea4b8105134d9135df8b567990b099a3040df8241b1fb4e8ad9c4e4.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "a1c1a8f6d253f92296d6843e081b757ad436206aa9f7737a27b5cf29f00aadc0", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f4e88d6145063cae6d7876457137d1c0fdbf9519a88cb3b01ca7d5ca27056621", "generation_seconds_batch": 23.199840966961347, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.178135871887207, "target_z": 0.3476562745374758, "ce": 5.831989288330078, "genuineness": 4.457139015197754, "identity": 0.18175113201141357, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/a1c1a8f6d253f92296d6843e081b757ad436206aa9f7737a27b5cf29f00aadc0.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "a1dd0f1a596587719bbfd9aee1df60745e7f4bf89ad8534ce50becc7f69632a3", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c60d7cbc4b0df116ddebecce4ab07c7a1fffa96ce82826c2177aa1d7a5f0421b", "generation_seconds_batch": 18.236230348004028, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4294402599334717, "target_z": 0.1269531045521035, "ce": 6.23706579208374, "genuineness": 4.469690799713135, "identity": 0.6739763021469116, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/a1dd0f1a596587719bbfd9aee1df60745e7f4bf89ad8534ce50becc7f69632a3.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "a2398ffe7d8fefb1cb7f3baeb18afecb20cc527409d20d2c7dc9b0876dffad9f", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f824f75e5ba4c2a2e508fc8fa3336e08499138598bde3a050cfa55c87b7b43f3", "generation_seconds_batch": 16.826629890943877, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0973048210144043, "target_z": 1.2968750327166343, "ce": 5.816103935241699, "genuineness": 4.758388519287109, "identity": 0.8604673147201538, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/a2398ffe7d8fefb1cb7f3baeb18afecb20cc527409d20d2c7dc9b0876dffad9f.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "a261667bdf44663e994956424c3dbf3404f012fcd0ebfbafee17bdde45abc2cc", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "99737a3e4210db372d8a6d4b0b179003d0886530566776147822cf8061722a76", "generation_seconds_batch": 16.80170062603429, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8230762481689453, "target_z": -0.21875001308665376, "ce": 5.935244083404541, "genuineness": 4.331617832183838, "identity": 0.6668317317962646, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/a261667bdf44663e994956424c3dbf3404f012fcd0ebfbafee17bdde45abc2cc.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "a2a4cd947a03d15c268cb20ae562c49012ffe6daa636240f0bb7d21756a2bdc5", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0e3660433de7df3a2c5111569369aac92ca04f12e2d98bc3241190fea05fd468", "generation_seconds_batch": 22.650859603076242, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1247615814208984, "target_z": 0.39453120256088015, "ce": 6.046441555023193, "genuineness": 4.494794845581055, "identity": 0.7344943284988403, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/a2a4cd947a03d15c268cb20ae562c49012ffe6daa636240f0bb7d21756a2bdc5.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": "a3114292df0a64d88ca19bb5977b90804c5a4e07862655cc6b1ba4b7abd37195", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "bf5b943cd1cd330fd0a5ca2fe68fbc227238cf9090dde3daac076aa3c20d99b6", "generation_seconds_batch": 20.011200624052435, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5242998600006104, "target_z": 0.9218749803700194, "ce": 5.506340026855469, "genuineness": 4.7081804275512695, "identity": 0.7691653966903687, "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/vn_STNC_reg_low/a3114292df0a64d88ca19bb5977b90804c5a4e07862655cc6b1ba4b7abd37195.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "a388e588f94be1b9d694e27ddb40c6a928b83a9a6a42b52082681c4bf3e6ed86", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "af1a25e166183122e0c67ab9ed736e0baa60e2d3a07991b1f6c2c6502f5e1581", "generation_seconds_batch": 22.96280747302808, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.490598440170288, "target_z": 0.07324212738318434, "ce": 6.070269584655762, "genuineness": 4.457139015197754, "identity": 0.6667947769165039, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/a388e588f94be1b9d694e27ddb40c6a928b83a9a6a42b52082681c4bf3e6ed86.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "a3c49ca2cf6d29c527805de130c3ae08b445333c1c8626ba1273c7fb3b7aff2a", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "cae8b4cbf59281aed52e88e40c0ff467239a3bb6d81b92d51060f237ae4d7a1a", "generation_seconds_batch": 17.273282513022423, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.848224401473999, "target_z": 1.515625045803288, "ce": 5.744620323181152, "genuineness": 4.720732688903809, "identity": 0.2995278239250183, "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/vn_STNC_reg_low/a3c49ca2cf6d29c527805de130c3ae08b445333c1c8626ba1273c7fb3b7aff2a.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "a3d944ed9bebe5b431b495e27bd12ec813f9a816412e2dbd22e9617c9ec2b00d", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d4acd840b8475447d9f679d558c3174e8a98266db53877c16d0a18e96790e47c", "generation_seconds_batch": 25.187210169970058, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9549732208251953, "target_z": 1.4218749803700195, "ce": 5.669164657592773, "genuineness": 4.507347106933594, "identity": 0.5962808728218079, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/a3d944ed9bebe5b431b495e27bd12ec813f9a816412e2dbd22e9617c9ec2b00d.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "a3f32e385837fd79c7a5181a403cd293fe9761a4e995a93da088836840a28055", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "502622697d39216ff1527ad8b1b0da3f1d3dffd20230e41ac06711a8439e94c1", "generation_seconds_batch": 22.393496081000194, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.07138729095459, "target_z": 0.4414061305842845, "ce": 6.09409761428833, "genuineness": 4.482243061065674, "identity": 0.7881429195404053, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/a3f32e385837fd79c7a5181a403cd293fe9761a4e995a93da088836840a28055.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "a419c0f35ac8d2e696f7736c22333d522643a4e661265dcd378d2e344281ba00", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "55cca22396bd6e49218938e94bcb2b1cf3223e9c99bc1c8394cd8f442f2ea0ba", "generation_seconds_batch": 16.177126638009213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1136417388916016, "target_z": 0.4042969870544727, "ce": 6.205295085906982, "genuineness": 4.482243061065674, "identity": 0.7297422885894775, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/a419c0f35ac8d2e696f7736c22333d522643a4e661265dcd378d2e344281ba00.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": "a4676ce38ecf806b2b7c09cbd797658b7729ce0618d9725f553552ee2f77a5a7", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "83bad3e2acc12af118326c831b2fd92e9284b5f4bd980fd2c6ee2275e1e1de8b", "generation_seconds_batch": 20.65967797394842, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0973048210144043, "target_z": 1.2968750327166343, "ce": 4.966238498687744, "genuineness": 4.532451152801514, "identity": 0.6457469463348389, "hypothesis": "J'ai écrit le le numéro sur l'enveloppe, lis le bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/a4676ce38ecf806b2b7c09cbd797658b7729ce0618d9725f553552ee2f77a5a7.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "a4a84a1b5901eecb93dac428b74ca900ec2d2477d450946a31299dd7a203b6cb", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "80a8ff657d055858441195d26347acceabc94de0054dbf660ee4fb2a8b65206c", "generation_seconds_batch": 22.229876670986414, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5154041051864624, "target_z": 0.9296875032716634, "ce": 5.959072113037109, "genuineness": 4.821148872375488, "identity": 0.5642869472503662, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/a4a84a1b5901eecb93dac428b74ca900ec2d2477d450946a31299dd7a203b6cb.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "a4ab3bcf34dfdb42a3b411bf267f60626c2d44af730fb38d82a35ac696ac8df9", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a09bd563416f89ece8a730829621b0df6a0aa663bf9aa732fc6b0ddb97046369", "generation_seconds_batch": 16.365986089920625, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.507833957672119, "target_z": 0.05810537080457585, "ce": 6.173524379730225, "genuineness": 4.532451152801514, "identity": 0.6928489208221436, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer ik 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/vn_STNC_reg_low/a4ab3bcf34dfdb42a3b411bf267f60626c2d44af730fb38d82a35ac696ac8df9.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "a4b310d1e3ad95944f7048c8b80f2bced668490ade9ca89c6094a61a279151b8", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c872d1f39a5fe548a74c8853cd4c9ff85ade74660f0afe6ab5f983edc148c447", "generation_seconds_batch": 16.978285963996314, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0758349895477295, "target_z": 0.4375000261733075, "ce": 6.300607204437256, "genuineness": 4.532451152801514, "identity": 0.4722454249858856, "hypothesis": "The 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/vn_STNC_reg_low/a4b310d1e3ad95944f7048c8b80f2bced668490ade9ca89c6094a61a279151b8.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "a4f26106a317bc473e3f82ddb31c282a09bc7170f2a356e622bf5dcb8bd5aeb7", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "64a17acbe0f52d01f54e1824c0ba503117bd7512b0a6725f1ef3126538e3dc42", "generation_seconds_batch": 18.62882692704443, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3960812091827393, "target_z": 0.15625003925996125, "ce": 6.022613525390625, "genuineness": 4.645420074462891, "identity": 0.7760090827941895, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/a4f26106a317bc473e3f82ddb31c282a09bc7170f2a356e622bf5dcb8bd5aeb7.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": "a5113988470e04cc971d6382d31507ec09b79e58433e928393319ab9d24d5497", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The road turns left behind the bridge.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f82d1a01d8100b181e1773fe88157ee7b2717949d199af91b99da125eddd7183", "generation_seconds_batch": 17.330610017990693, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1002984046936035, "target_z": 0.41601550967386386, "ce": 5.9431867599487305, "genuineness": 4.2374773025512695, "identity": 0.5492706298828125, "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/vn_STNC_reg_low/a5113988470e04cc971d6382d31507ec09b79e58433e928393319ab9d24d5497.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "a54810bdbe6adf305416673df5761d004d6618758db48560e9fe069c77375308", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "196902006b12eedb940913adee41aff6d655d58e36508ae339c94c7c3b48dca6", "generation_seconds_batch": 21.482714388053864, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6666314601898193, "target_z": 0.7968750327166344, "ce": 5.959072113037109, "genuineness": 4.770940780639648, "identity": 0.6631588935852051, "hypothesis": "La puerta sigue abierta. He dejado tu ahrigo 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/vn_STNC_reg_low/a54810bdbe6adf305416673df5761d004d6618758db48560e9fe069c77375308.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "a589dcb0dbb30c69f21e55ccde08ca267b8a24755d3cb4d32634966adf3128b8", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "de89e46a063e7f90b2d3348cdbfd75cf302527c506eff8ff04addf918ca0d674", "generation_seconds_batch": 15.10150136006996, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0091171264648438, "target_z": 0.496093686202563, "ce": 5.8875885009765625, "genuineness": 4.6705241203308105, "identity": 0.7792532444000244, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/a589dcb0dbb30c69f21e55ccde08ca267b8a24755d3cb4d32634966adf3128b8.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "a5af3a8678c7d6d5d48a2a62bf1b0c1386bc8658677b437d6632aa4410609a6b", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1717e5840101645e6bd188be21d13c98eda7e9bbaba13719bdb41f8302ced655", "generation_seconds_batch": 17.462195041007362, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4016411304473877, "target_z": 0.15136714701316498, "ce": 6.1894097328186035, "genuineness": 4.607763767242432, "identity": 0.49202316999435425, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/a5af3a8678c7d6d5d48a2a62bf1b0c1386bc8658677b437d6632aa4410609a6b.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "a65d04169840737832f9b7cde12a8f5ff8789c9b39639727a0154d7ea152418e", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e75858d6ecc979ddf9a3667878cd2a66db6ac3c340da8d5b3f7532f2d0a1e5fd", "generation_seconds_batch": 22.9115524640074, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6168065071105957, "target_z": -0.037597642140951425, "ce": 6.356205940246582, "genuineness": 4.607763767242432, "identity": 0.528741717338562, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/a65d04169840737832f9b7cde12a8f5ff8789c9b39639727a0154d7ea152418e.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "a67461248c056294f8a65e798f25f17541595d2376bb0ee63230c5e75c83e65d", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b59729567bc7ed6351d6ab72844fb9b9267bc294c0bb99084ee33618848da7aa", "generation_seconds_batch": 22.53665775002446, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8267545700073242, "target_z": 0.6562500392599613, "ce": 5.998785495758057, "genuineness": 4.425758361816406, "identity": 0.6026151180267334, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/a67461248c056294f8a65e798f25f17541595d2376bb0ee63230c5e75c83e65d.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "a6e03d79fbe84596414fd97cada18698b99b8186c6597f5d9eaa0d6452b52c25", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9cdda819de3997c7d2354f7796803884822e14764eea584ce5d2960dc282de96", "generation_seconds_batch": 21.541966571006924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.857120156288147, "target_z": 1.5078125229016441, "ce": 5.7684478759765625, "genuineness": 4.394378662109375, "identity": 0.41186076402664185, "hypothesis": "I found the little box. It was beside the old window. We can l 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/vn_STNC_reg_low/a6e03d79fbe84596414fd97cada18698b99b8186c6597f5d9eaa0d6452b52c25.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": "a710b4568771715a51b0f168a6e98ca6accb89f6db27aeeb133221c4c75bf3a0", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "26b075bf01e4f4ce3d5dfe18ecd3edf3dadff3fc840fd2c2555e7003466a8e84", "generation_seconds_batch": 20.422780020046048, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8660159111022949, "target_z": 1.5, "ce": 5.597680568695068, "genuineness": 4.7081804275512695, "identity": 0.7305507659912109, "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/vn_STNC_reg_low/a710b4568771715a51b0f168a6e98ca6accb89f6db27aeeb133221c4c75bf3a0.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "a7504df736fb1b8fb1a51e5a3b18b47f4a74cf18f7dabfc042b578999b50421b", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2616e7cec528b9c1bb3dc95ef7bde6078b98fcffd9986531af2a165b3ae8d414", "generation_seconds_batch": 16.308418160071597, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7948501110076904, "target_z": 1.5624999738266925, "ce": 5.502368927001953, "genuineness": 4.8588056564331055, "identity": 0.713897705078125, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise Dis-moi, quand tu es prêt", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/a7504df736fb1b8fb1a51e5a3b18b47f4a74cf18f7dabfc042b578999b50421b.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "a7a6d7e2f98d6126d3b0a29567314a41e1b37935eac2aee34fa9e082f29980cb", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "67c30510aca136d5097bcdc04cf250572d11b2f9cced85ce430ece49a5b01604", "generation_seconds_batch": 22.424719608039595, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6933186054229736, "target_z": 0.7734375687049322, "ce": 5.911416053771973, "genuineness": 4.557555198669434, "identity": 0.6132898330688477, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/a7a6d7e2f98d6126d3b0a29567314a41e1b37935eac2aee34fa9e082f29980cb.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "a7bab66126febf4c1fd1a16947b045752c2bfceaca7382136322131985e94af9", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e5c09016f69aae0dce55af7d59990f062a77f9b30e01b49709df2bbfce02d534", "generation_seconds_batch": 15.10150136006996, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1069700717926025, "target_z": 0.4101562483641683, "ce": 5.95112943649292, "genuineness": 4.557555198669434, "identity": 0.7357528805732727, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/a7bab66126febf4c1fd1a16947b045752c2bfceaca7382136322131985e94af9.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "a81ce30e5c8af793f0d5fe23d4755ee90568c9330176917775a0f160115c8579", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "52e0ef3c04e151ee92538bb6ebb738099cb7361cbcebf6d66dbd4c4a3478e1fe", "generation_seconds_batch": 16.49756517598871, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.511725664138794, "target_z": 0.054687555618278434, "ce": 6.284721851348877, "genuineness": 4.432034492492676, "identity": 0.2565913796424866, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/a81ce30e5c8af793f0d5fe23d4755ee90568c9330176917775a0f160115c8579.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": "a860c85f038417054040b8aaac68cd6684778c82c9aea25aeba34f9acef4ffb7", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d972f115ab309f0d1262b5c5ff1d6887fa79f074e64c250a8f18a754a8df7ac4", "generation_seconds_batch": 20.242094643996097, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5687785148620605, "target_z": 0.8828124705550291, "ce": 5.605623245239258, "genuineness": 4.7332844734191895, "identity": 0.8485521674156189, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, Lisi, 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/vn_STNC_reg_low/a860c85f038417054040b8aaac68cd6684778c82c9aea25aeba34f9acef4ffb7.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "ad1da03da04458b5058aedb79efd4f609d398a56e3d507eaf6a8987d5fc22024", "generation_seconds_batch": 17.001083524082787, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.125530958175659, "target_z": -0.4843749541967119, "ce": 6.117925643920898, "genuineness": 4.457139015197754, "identity": 0.6498521566390991, "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/vn_STNC_reg_low/a8bd001b1fcafd2f49b65d73e6ed12567e4c9664909a4d08700c9febfd977540.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "a90b5210d6e72c8fc5cdb31234e47dbd9a4c4079207ebacb923166fe7ae780bb", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0f4f95dd9602d95f22d1b5ab59d4f9dab7acdc6962d02e28a912605f213e7967", "generation_seconds_batch": 16.543650471023284, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.213718891143799, "target_z": 0.31640618293089956, "ce": 5.998785495758057, "genuineness": 4.507347106933594, "identity": 0.6275104284286499, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/a90b5210d6e72c8fc5cdb31234e47dbd9a4c4079207ebacb923166fe7ae780bb.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "211c337be48e890d749739f150be3e7818f7481d8c61fe75416459e7e599a603", "generation_seconds_batch": 15.356928450986743, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7152154445648193, "target_z": -0.12402336920402575, "ce": 6.181467056274414, "genuineness": 4.595211505889893, "identity": 0.6794134378433228, "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/vn_STNC_reg_low/a94b7d5da78ff217f8e5a4f7ae4520900bef0ad764d0e4ee898e69de10c28aea.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "a94eb77e108f1dc2527dd8cb54a7bc5ea4948a2d37dd20c30f06550238c6b97d", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4990c3c0bbdf4c908164d9dba80664b79d1e8b530d8b64edec97f99b26ed938c", "generation_seconds_batch": 21.715229570982046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.648840069770813, "target_z": 0.8124999738266925, "ce": 5.760505676269531, "genuineness": 4.833701133728027, "identity": 0.7702552080154419, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand 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/vn_STNC_reg_low/a94eb77e108f1dc2527dd8cb54a7bc5ea4948a2d37dd20c30f06550238c6b97d.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": "a97c3eeaa0e471a70838321f2354cdc3484400123541b322127ac89023d89773", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Escribí el número en el sobre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7816e902ccd6f3b218ccf4e9d05b7ed4b85b3a8d391af117fd2c78815d0533b9", "generation_seconds_batch": 16.13305504701566, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.011341094970703, "target_z": 0.49414052930384444, "ce": 5.5659098625183105, "genuineness": 4.545003414154053, "identity": 0.5304257869720459, "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/vn_STNC_reg_low/a97c3eeaa0e471a70838321f2354cdc3484400123541b322127ac89023d89773.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "aa2cc1fde35ae8a0703ff47a78b7c7b9207946f3cc964da21055f9948656796a", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "da80a998e1523842482620a939c5ff01ae4232d3349d7969b2bfb00196fe53bc", "generation_seconds_batch": 21.185890013934113, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.16479229927063, "target_z": 0.3593750065433269, "ce": 6.038498878479004, "genuineness": 4.519899368286133, "identity": 0.6906147599220276, "hypothesis": "La porte est encore ouverte. J'ai posé ton j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.16666666666666666, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/aa2cc1fde35ae8a0703ff47a78b7c7b9207946f3cc964da21055f9948656796a.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "aaa25c6e7f3f62d501419781f697ca26735dffb3355dfa355b970b996011ddc7", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "52f3f45e347c5aea90761a74fb2196aec7c06388e6d5df39d1da14f8395c343f", "generation_seconds_batch": 17.107407317962497, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1203136444091797, "target_z": 0.3984375163583172, "ce": 5.919358730316162, "genuineness": 4.400654315948486, "identity": 0.7360001802444458, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/aaa25c6e7f3f62d501419781f697ca26735dffb3355dfa355b970b996011ddc7.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "aaa5afe7ec3d3ce9d957e3451d0f2b7718d4b27eeb8f5ce0752bfeabe60e7ece", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a8066903542e3bb3ee1551f868ab2cdfef8c6c232015099754815c8d0b7f12ec", "generation_seconds_batch": 22.709262764896266, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2396365404129028, "target_z": 1.1718749803700195, "ce": 5.704906940460205, "genuineness": 4.287685871124268, "identity": 0.5911905765533447, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/aaa5afe7ec3d3ce9d957e3451d0f2b7718d4b27eeb8f5ce0752bfeabe60e7ece.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "aab74eda2af02779c788e83543f512758027a4f49a5d0df44d627ef6dfae2173", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "830efcd6d13907c59e4209f2eda9d311b7edf2a4e8186f44ab75fb4fee04422d", "generation_seconds_batch": 23.719138300977647, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6755272150039673, "target_z": 0.7890625098149903, "ce": 6.046441555023193, "genuineness": 4.507347106933594, "identity": 0.6933842897415161, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/aab74eda2af02779c788e83543f512758027a4f49a5d0df44d627ef6dfae2173.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": "aac316eee57983163b153cf5abd3f2b672bc73a0f1a0bf3e7c71ba2dad05e468", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b7b141e55c3b06b5524ee99f56a73d70bb72797f7656cade15c2f27969659cdc", "generation_seconds_batch": 20.5348345200764, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9371817111968994, "target_z": 1.4375000261733075, "ce": 5.458683967590332, "genuineness": 4.720732688903809, "identity": 0.7448434829711914, "hypothesis": "Tu mensaje Wolf después de que me fuera, lo leí dos veces y guardé el teléfono, dime cuándo podemos hablar.", "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/vn_STNC_reg_low/aac316eee57983163b153cf5abd3f2b672bc73a0f1a0bf3e7c71ba2dad05e468.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "aad7e4fb82c36a9b581874363657d1964fa406645a2db4649007d01fe6c9ebf6", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1751ba47d72dff011f9465bbd02eb94d3a3dea0d66d2040eb9caa17754df846c", "generation_seconds_batch": 16.972046782029793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5331956148147583, "target_z": 0.9140624574683753, "ce": 5.80021858215332, "genuineness": 4.7081804275512695, "identity": 0.5283578634262085, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/aad7e4fb82c36a9b581874363657d1964fa406645a2db4649007d01fe6c9ebf6.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": "ab1ce5abc9e337fa1c3b37db7a333b4d28f126bba869a51b5e9f77133bb04021", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "abdb52fb6d8678e4a7c456749cd96a205ef595a1427b292ae3e572962332649d", "generation_seconds_batch": 16.77914552600123, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7778280973434448, "target_z": 0.6992187581791586, "ce": 5.696964263916016, "genuineness": 4.532451152801514, "identity": 0.623757004737854, "hypothesis": "Las llaves están en la mesa de la cocina, miren el cajón y luego junto a la ventana, llévatelas cuando te vayas.", "wer": 0.08695652173913043, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ab1ce5abc9e337fa1c3b37db7a333b4d28f126bba869a51b5e9f77133bb04021.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "ab55fdabf505f38b0517f3842f92a39ceea95c9e39c60897eb37e533fc8a298d", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ed37c5e9843aa1339b5cd663df53a50c974ff63e1ca0be29fe033f75612b0c1c", "generation_seconds_batch": 23.948627518955618, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1759119033813477, "target_z": 0.3496094314361943, "ce": 5.919358730316162, "genuineness": 4.494794845581055, "identity": 0.7338586449623108, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ab55fdabf505f38b0517f3842f92a39ceea95c9e39c60897eb37e533fc8a298d.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "944482d32489724437460757bf91f8307d50defc1ec04d8d6ae7e5d6a71bf305", "generation_seconds_batch": 18.000636689015664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.081052303314209, "target_z": -0.4453124443817216, "ce": 6.427689552307129, "genuineness": 4.369274139404297, "identity": 0.412036657333374, "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/vn_STNC_reg_low/ab7a94299f968552a7b966fa3391724b652d8e20483888a7767196f29a7f4a69.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "aba30e83e624a41c24c62528e8a5ac7f34962c7247eeb8d093b072ee8030201b", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2caec7a7619731c06033f248821918445df4412b568062549fd5d0392fbcc4c0", "generation_seconds_batch": 22.10517268104013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.577674150466919, "target_z": 0.875000052346615, "ce": 5.71682071685791, "genuineness": 4.507347106933594, "identity": 0.5703991055488586, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/aba30e83e624a41c24c62528e8a5ac7f34962c7247eeb8d093b072ee8030201b.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "abaeb510636e0d76beede7ff3537bff40927c3f7feed5869d6b8f9fa49846e58", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ff88acb3489534a3bb31d480b35be47f937a84d252440d96c838cfd8975b277d", "generation_seconds_batch": 17.562141527072527, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8126415014266968, "target_z": 1.5468750327166343, "ce": 5.617537498474121, "genuineness": 4.833701133728027, "identity": 0.5964594483375549, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/abaeb510636e0d76beede7ff3537bff40927c3f7feed5869d6b8f9fa49846e58.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "abf8c447f793daf9dfb07532c682825fc93c3371b0a0ae2050942ba44c81a67c", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "046e45df647015ebfdea0ca1cf4efcbee66d6bd86f9d1189a0a2c409da7f31ad", "generation_seconds_batch": 20.37721604993567, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3730725049972534, "target_z": 1.0546874509250483, "ce": 5.824047088623047, "genuineness": 4.7081804275512695, "identity": 0.7538723945617676, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/abf8c447f793daf9dfb07532c682825fc93c3371b0a0ae2050942ba44c81a67c.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "acd55489b598180256fabb90794d1a9c0d3c8b7bace874def0ac11e7e2687edb", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b491ea2d4f9d891df57e4ef01688c14fa7826298087d1a8cb19b1663ee65dfb2", "generation_seconds_batch": 16.390651933965273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.592343330383301, "target_z": -0.01611333502796775, "ce": 6.260893821716309, "genuineness": 4.570107460021973, "identity": 0.5439156293869019, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/acd55489b598180256fabb90794d1a9c0d3c8b7bace874def0ac11e7e2687edb.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": "ad06768d911527befb09660c7b42141a726e192d12d6faefada3c66abad98a65", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "662dae5f1cac7a2a03f7526c2f83d34dcd1c0fc692e8e49827e29b30c7816821", "generation_seconds_batch": 20.78684511699248, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.079513430595398, "target_z": 1.3124999738266925, "ce": 5.561938762664795, "genuineness": 4.595211505889893, "identity": 0.7489820122718811, "hypothesis": "La route tourne à gauche après le pont, continue jusqu'à la petite barrière en bois. Je je t'attendrai de l'autre côté.", "wer": 0.05, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ad06768d911527befb09660c7b42141a726e192d12d6faefada3c66abad98a65.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "ad5b8a56920d6444c1cedf037a5bc4795dcee87d05f421cc62183c924362e59a", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8d5941c7bdacfb8091110cd562888fefa77776aa9deaa5931b4c8978aab2fbca", "generation_seconds_batch": 22.56462784006726, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.417551040649414, "target_z": 1.015625045803288, "ce": 5.621508598327637, "genuineness": 4.758388519287109, "identity": 0.780138373374939, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ad5b8a56920d6444c1cedf037a5bc4795dcee87d05f421cc62183c924362e59a.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "cedcd601d71ef94d59c83db39bc62726634d0d42fb22b4357678f0873927ffc4", "generation_seconds_batch": 17.95327451103367, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0721566677093506, "target_z": -0.4375000261733075, "ce": 6.141753673553467, "genuineness": 4.432034492492676, "identity": 0.3332451581954956, "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/vn_STNC_reg_low/adcc0121d6c11de6953f8243f84ad890832f75283c96342c1db246fddc443438.mp3" }, { "id": "heldout-en-2", "language": "en", "sentences": [ "The road turns left behind the bridge.", "Keep going... until you reach the small wooden gate.", "I will meet you on the other side." ], "text": "The road turns left behind the bridge. Keep going... until you reach the small wooden gate. I will meet you on the other side.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-2", "key": "ae3130a5db36c2790a31e23135fde257111c56fa9c5b8061c1f3e51a8cb61258", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "844e8ae5a674e9b2747c7f81f68e5616196ad7828bfcd7644dbb803d15f4b45d", "generation_seconds_batch": 15.173490614979528, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.724111318588257, "target_z": -0.1318359967988998, "ce": 6.372091293334961, "genuineness": 4.13078498840332, "identity": 0.40461835265159607, "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/vn_STNC_reg_low/ae3130a5db36c2790a31e23135fde257111c56fa9c5b8061c1f3e51a8cb61258.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "aea30337fe2e21765b2484ef53eb7e8dd5186a0ba16fdb0d1e966c5b3a0f6f63", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "717533a3b56314c27f84c1226632f343acb66b989b194a213c97d4b25e3a42cc", "generation_seconds_batch": 16.515220660017803, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6844229698181152, "target_z": 0.7812499869133462, "ce": 6.070269584655762, "genuineness": 4.6956281661987305, "identity": 0.7131931781768799, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/aea30337fe2e21765b2484ef53eb7e8dd5186a0ba16fdb0d1e966c5b3a0f6f63.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "aededf16d2eccc914be43dea1cb245f04fd8d2f351b3eeb6941b56918d71326e", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "19de659b984ab062622e8a0cdd302c6976ab4365b2e5bcc5767c5500fdc50677", "generation_seconds_batch": 22.986359643982723, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0343499183654785, "target_z": -0.4042969870544727, "ce": 6.443574905395508, "genuineness": 4.419482707977295, "identity": 0.5836807489395142, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/aededf16d2eccc914be43dea1cb245f04fd8d2f351b3eeb6941b56918d71326e.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": "af23c52e32235d286e61dc3cd942b848ecf514a7578147c247af428c99e5eccd", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "ee9ae7de1659dbc427928aa65d8d33760e2ccbcc52baac26fe55ff0885f98d9e", "generation_seconds_batch": 17.604474638937972, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7555887699127197, "target_z": 0.7187500130866538, "ce": 5.816103935241699, "genuineness": 4.7834930419921875, "identity": 0.77528977394104, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/af23c52e32235d286e61dc3cd942b848ecf514a7578147c247af428c99e5eccd.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": "af31dc901d82a22108368b350190814d89e8f8e52153aa56735a92fc024df51c", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7aadafe2d9185018ae0132c62dbc6eaf7ea78b6a33b5cab06a4140523afe3f09", "generation_seconds_batch": 16.80195873102639, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1862621307373047, "target_z": 1.2187500130866538, "ce": 6.086154937744141, "genuineness": 4.607763767242432, "identity": 0.44470471143722534, "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/vn_STNC_reg_low/af31dc901d82a22108368b350190814d89e8f8e52153aa56735a92fc024df51c.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "af397ffd78529d3631c52cf68bf9a7440b682908e46822945fc210e5ccf096c0", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7a4f0b1a7905db07315c1622e617547a4bfbed2d73ce287f29f965651fe67a71", "generation_seconds_batch": 22.851364523055963, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7414757013320923, "target_z": 1.6093750065433268, "ce": 5.593709468841553, "genuineness": 4.821148872375488, "identity": 0.6271476745605469, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise, dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/af397ffd78529d3631c52cf68bf9a7440b682908e46822945fc210e5ccf096c0.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "af4ccc71dd9493a7f7981fc46385ac03d5fb093de0ca854a1b43045dda7485b8", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "484d1811c9c70a4d99cf6f13146fa71fa275348da98b812634e9b599e7356b68", "generation_seconds_batch": 22.78277089993935, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6169354915618896, "target_z": 1.7187500130866538, "ce": 5.43088436126709, "genuineness": 4.645420074462891, "identity": 0.8460190296173096, "hypothesis": "J'ai trouvé ma petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/af4ccc71dd9493a7f7981fc46385ac03d5fb093de0ca854a1b43045dda7485b8.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": "af64ffc1835e38435745512129af44269958a92e6db4aca7a094539556393f2a", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4d9b8741ac400e6932bc297bd6df8ba3505374302bf5e28f67d182c186169047", "generation_seconds_batch": 20.663872670033015, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.862337589263916, "target_z": 0.624999947653385, "ce": 5.792276382446289, "genuineness": 4.331617832183838, "identity": 0.5540755987167358, "hypothesis": "Ik schreef het nummer op de enveloppe, 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/vn_STNC_reg_low/af64ffc1835e38435745512129af44269958a92e6db4aca7a094539556393f2a.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "af7aa8e5736029832de81dc07d865bbaa5ec598f9c6cf28646fa60bd3f03c98e", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "17fa513581316a9bac5e8433b600082da0a0614d94254cfaa95dc055b15e318f", "generation_seconds_batch": 23.426658542011864, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8838074207305908, "target_z": 1.484374954196712, "ce": 5.1628193855285645, "genuineness": 4.570107460021973, "identity": 0.7449906468391418, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/af7aa8e5736029832de81dc07d865bbaa5ec598f9c6cf28646fa60bd3f03c98e.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "af8641c8fed776e950e43390f2ad3e97bb9e50eb7dffe4aef97d34ff3964e353", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d10d756de94364381e8a2639c246f12d6b646c4e7109e80e96f01eb68d7aa24a", "generation_seconds_batch": 25.187210169970058, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1328877210617065, "target_z": 1.265625045803288, "ce": 5.756534099578857, "genuineness": 4.444586753845215, "identity": 0.5866549015045166, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/af8641c8fed776e950e43390f2ad3e97bb9e50eb7dffe4aef97d34ff3964e353.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "af8d25aa1b5c88578b28301f71a78bf216e028e909e64d22413478f46d2038f6", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7ddb5aa25953eba837354907d247917e53aae980fcfd47f0806cfb23fb3e49bd", "generation_seconds_batch": 17.743624301976524, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7141034603118896, "target_z": -0.12304679075466649, "ce": 6.133810997009277, "genuineness": 4.7081804275512695, "identity": 0.6919269561767578, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/af8d25aa1b5c88578b28301f71a78bf216e028e909e64d22413478f46d2038f6.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": "af98e267e8d94c6edab143b98fab585a68afecdbb68c71fb9c3bec89f92810e3", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The keys are on the kitchen table.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4451547ab33c03ef193dd0a7a342d2aa5af955ce5fd38263ce43de26a6d352da", "generation_seconds_batch": 23.769224004936405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3997596502304077, "target_z": 1.0312499869133462, "ce": 6.086154937744141, "genuineness": 4.595211505889893, "identity": 0.47581008076667786, "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/vn_STNC_reg_low/af98e267e8d94c6edab143b98fab585a68afecdbb68c71fb9c3bec89f92810e3.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "b007c501dc6ead90604d8d7de24c5e54aac87cd9ea139c586cdf4f2fc47f635d", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "db35c55aecefb767574d4694274ee770ef455b0034b71f770d24b2093bc50fd0", "generation_seconds_batch": 27.057254050974734, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0891785621643066, "target_z": 0.4257812941674564, "ce": 5.863759994506836, "genuineness": 4.419482707977295, "identity": 0.7469607591629028, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/b007c501dc6ead90604d8d7de24c5e54aac87cd9ea139c586cdf4f2fc47f635d.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "b080a1ad2d6ca799c4ac875ac7e932e9b3ce086537c015d7868ab595886dc990", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5e193d679c6b594d4c816018421cc3e23a03fe3c94a9671edd9e1459548621a9", "generation_seconds_batch": 16.88433393498417, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9965429306030273, "target_z": -0.371093738549178, "ce": 6.324435234069824, "genuineness": 4.319066047668457, "identity": 0.5196982622146606, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/b080a1ad2d6ca799c4ac875ac7e932e9b3ce086537c015d7868ab595886dc990.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "b08e9f7dac8261a732506fd96ec5c0e2e1e6d4027602de8248c6b01001ff22cb", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ca1b1bdacf635143561b8683a3b98579df57aca2ed4a38096d5645ce1d997d19", "generation_seconds_batch": 17.208157390006818, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.61770498752594, "target_z": 0.8398437516358317, "ce": 5.911416053771973, "genuineness": 4.18099308013916, "identity": 0.49815306067466736, "hypothesis": "The 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/vn_STNC_reg_low/b08e9f7dac8261a732506fd96ec5c0e2e1e6d4027602de8248c6b01001ff22cb.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "b0fd4c0e2836aa9f7d9636119d8e89d624707115a38db04dd14a62b653a1450f", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eaebb4fe053a1313c2d670b61a7a81f1eabbd1b0fcda323d90bba18fcaf79b90", "generation_seconds_batch": 16.901126254932024, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6140265464782715, "target_z": -0.035156196017553286, "ce": 6.205295085906982, "genuineness": 4.570107460021973, "identity": 0.6559663414955139, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/b0fd4c0e2836aa9f7d9636119d8e89d624707115a38db04dd14a62b653a1450f.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "b124791cd8a6db3b745b2a277eef2c16428573712c9030f7db4a35a958712b43", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ae4a0ce4bd0f403148026044c55dec42bdb758f530f9c915e5e9da2299fe53a0", "generation_seconds_batch": 23.826806705095805, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8393287658691406, "target_z": 1.523437464011702, "ce": 5.645336627960205, "genuineness": 4.996878147125244, "identity": 0.8112477660179138, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/b124791cd8a6db3b745b2a277eef2c16428573712c9030f7db4a35a958712b43.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "b15c2dbd09ec8cab0c089a30c05aa3f2a6d393be3ae2ccde157f34bf6ffa5cd2", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "68c432cb67ea8b65e13700e97f644d1105ee3ab75824890e927b707c77be56cc", "generation_seconds_batch": 22.14416596200317, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9512948989868164, "target_z": 0.5468749280234044, "ce": 6.046441555023193, "genuineness": 4.632867813110352, "identity": 0.5520538687705994, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/b15c2dbd09ec8cab0c089a30c05aa3f2a6d393be3ae2ccde157f34bf6ffa5cd2.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "b160f7fba2afb9a776b3b44162aa919cf0833d81f98b1f7ba3c2626d8599c942", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "094456e1ffc2874842e969eab013e302e956b6ac2b21bd5f2c41198e06ed21fa", "generation_seconds_batch": 17.84198633604683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6901962757110596, "target_z": -0.10205077286636244, "ce": 6.109982967376709, "genuineness": 4.65797233581543, "identity": 0.80666184425354, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/b160f7fba2afb9a776b3b44162aa919cf0833d81f98b1f7ba3c2626d8599c942.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "b166bf97e2e66d363f26e518cecfaacc82b5cebfedab3e76cec6184c36d1c54b", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9221f8defb56a447a52bd09821907258441c7dda9a34507ad6e6697f4a22bb51", "generation_seconds_batch": 25.52765471499879, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7774856090545654, "target_z": -0.1787109248223042, "ce": 6.054384231567383, "genuineness": 4.469690799713135, "identity": 0.5911217927932739, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/b166bf97e2e66d363f26e518cecfaacc82b5cebfedab3e76cec6184c36d1c54b.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": "b179a25a42e14a0196add960eadd6a3c26725c828050d0d22f14fed9153cb651", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "da8f71d14225eeb041ae8ed68e740b64d1a3e049dde2b359844eb3bd85dde675", "generation_seconds_batch": 17.83537497301586, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6043614149093628, "target_z": 0.8515624836416829, "ce": 5.756534099578857, "genuineness": 4.595211505889893, "identity": 0.752676784992218, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suche dir in der Schublade, dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/b179a25a42e14a0196add960eadd6a3c26725c828050d0d22f14fed9153cb651.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "b2105100e82f2e2c75d71686db5c270b51b38852218640538c7623f7e4d2a898", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "11bb315ce257396c96380b306ca6a762d030a11e06863f8e332cf65eb6b3f3b1", "generation_seconds_batch": 22.696449199924245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2804367542266846, "target_z": 0.2578125229016441, "ce": 5.911416053771973, "genuineness": 4.582659721374512, "identity": 0.7574062347412109, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/b2105100e82f2e2c75d71686db5c270b51b38852218640538c7623f7e4d2a898.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "b233632f78b221b92b428928a06a4146652911895e8c68299a0736ca3d3f2d96", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "42e296b6329823e8815679f58ee4a373d2ad68a99f8b7bf6473b9430448fe7df", "generation_seconds_batch": 17.937688520993106, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.126985549926758, "target_z": 0.39257804566216165, "ce": 6.205295085906982, "genuineness": 4.796044826507568, "identity": 0.8238893747329712, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/b233632f78b221b92b428928a06a4146652911895e8c68299a0736ca3d3f2d96.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "b2e089dc56089aa71f2cc147cdad5e78e827c50bc2e62f92235955bd7f2126eb", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "76f5815eb8f4f049e1a39e6a835660826d868a17a0a52249859851f446d845d0", "generation_seconds_batch": 16.50804863893427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0246846675872803, "target_z": 0.4824217972979934, "ce": 5.855817794799805, "genuineness": 4.532451152801514, "identity": 0.633197009563446, "hypothesis": "De deur staat nog open en 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/vn_STNC_reg_low/b2e089dc56089aa71f2cc147cdad5e78e827c50bc2e62f92235955bd7f2126eb.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": "b2f18f7a53b73595231f86af026e173c7d97c28307484b15fe77f17f1513533a", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f5d97787c0bcd2d491c17b296e956cb2c772bc331f68293d2f10035161f9cbc5", "generation_seconds_batch": 20.663872670033015, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7066622972488403, "target_z": 0.761718732005851, "ce": 5.895530700683594, "genuineness": 4.413206577301025, "identity": 0.7463313341140747, "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/vn_STNC_reg_low/b2f18f7a53b73595231f86af026e173c7d97c28307484b15fe77f17f1513533a.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "b31813e8b7ceaf0564c567731fedc0d5a41e4934f0f0a11ca22835e752e8fd10", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fa47095745d9931af00dc1efb20b734f61ed1e434b8b27824a0aec84b62e8d5a", "generation_seconds_batch": 17.800279836985283, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3819681406021118, "target_z": 1.0468750327166343, "ce": 5.772419452667236, "genuineness": 4.607763767242432, "identity": 0.8499554991722107, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise, dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/b31813e8b7ceaf0564c567731fedc0d5a41e4934f0f0a11ca22835e752e8fd10.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "b3769a8d7d66c3ec7f49801827c3143e25ae1ef57388413d12f721a5d6bf7a21", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d8aa71bbbe55acaf202c838525365ee3924b58472a11e99c31796f206ecebccb", "generation_seconds_batch": 24.103616160107777, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8660159111022949, "target_z": 1.5, "ce": 5.669164657592773, "genuineness": 4.796044826507568, "identity": 0.74140864610672, "hypothesis": "La porte 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/vn_STNC_reg_low/b3769a8d7d66c3ec7f49801827c3143e25ae1ef57388413d12f721a5d6bf7a21.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "b421e39be372310697d7c380a886ea961c3d24dd727a9709c84b1584fd4e1da5", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4637a11e8bb92ad01f255c218127edd28708096578dbeb37e821ab6b715ecd67", "generation_seconds_batch": 23.382360587012954, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.386073589324951, "target_z": 0.16503903591773458, "ce": 6.173524379730225, "genuineness": 4.268857479095459, "identity": 0.3447694182395935, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/b421e39be372310697d7c380a886ea961c3d24dd727a9709c84b1584fd4e1da5.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "0f810d96bc50dbcbdf28469b0f0fdcc5a187b64319a95b0e77dd0b0314b366e3", "generation_seconds_batch": 17.42142223601695, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.582613468170166, "target_z": -0.007568273596074255, "ce": 6.181467056274414, "genuineness": 4.406930446624756, "identity": 0.579231858253479, "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/vn_STNC_reg_low/b429ad869b4f073af736e4e6510807542fc4f5324b0eb7f096c719d9002f4f71.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "b467da6c7fcef17e18d36255f230bfe1c2c52b9eee2ecddaf33a7216c9a26aa4", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d1f082697b1f5f845fac1be62bb677f00ad1f9603423fd64eef61bc7b2492534", "generation_seconds_batch": 22.229876670986414, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1328877210617065, "target_z": 1.265625045803288, "ce": 5.7406487464904785, "genuineness": 4.821148872375488, "identity": 0.5843048095703125, "hypothesis": "He encontrado la cajita. Estaba junto la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/b467da6c7fcef17e18d36255f230bfe1c2c52b9eee2ecddaf33a7216c9a26aa4.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "b47b671420e926ddb6d6066c272b3f3271ab81f809601296f7f71a935731ad89", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5aad5b4a50a9e634c9aa1bd0d5e0c1df2fd0c6cdb42246e0f79d635ecacd0079", "generation_seconds_batch": 23.26596711901948, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0091171264648438, "target_z": 0.496093686202563, "ce": 5.8875885009765625, "genuineness": 4.7834930419921875, "identity": 0.6960169076919556, "hypothesis": "He encontrado la cajita, estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/b47b671420e926ddb6d6066c272b3f3271ab81f809601296f7f71a935731ad89.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "b4849dad30d80826d3893451a67872b97a34247d098680adf98f2a1b7ba770ba", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ada85503f77c09be77c5031c4bf1798efe605f6b008eaa2026345fed1e060343", "generation_seconds_batch": 20.553295812918805, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2426300048828125, "target_z": 0.29101556202047885, "ce": 6.078212261199951, "genuineness": 4.6203155517578125, "identity": 0.5927029252052307, "hypothesis": "Ich hab die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/b4849dad30d80826d3893451a67872b97a34247d098680adf98f2a1b7ba770ba.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": "b4b60fd49dbfef93c5a78dcbd73c8dd691a1195ea6afa500cbb3468c9ff9299a", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6f9ec941d78fca5e645587896e0ed9556e6a088873193225755422b8c707b0bc", "generation_seconds_batch": 20.935181880020536, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4709254503250122, "target_z": 0.9687500130866538, "ce": 6.070269584655762, "genuineness": 4.7458367347717285, "identity": 0.7608627676963806, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich lasse sie zweimal. Dann legte ich das Handy weg und sag mir, wann wir reden können.", "wer": 0.08333333333333333, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/b4b60fd49dbfef93c5a78dcbd73c8dd691a1195ea6afa500cbb3468c9ff9299a.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "b4c95f3d803625619bea6d8f6a7fdba7068b4d03dadcf6bf8564b2d34f5d3f61", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a6eb03bf115d996053544f59663ebb8161e64cb1dc21f70ce6ec78e8f5bb599a", "generation_seconds_batch": 21.911886514048092, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3285938501358032, "target_z": 1.0937499607400387, "ce": 5.935244083404541, "genuineness": 4.796044826507568, "identity": 0.7484108209609985, "hypothesis": "J'ai trouvé la la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/b4c95f3d803625619bea6d8f6a7fdba7068b4d03dadcf6bf8564b2d34f5d3f61.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "b4d2f9ad74156733ea2cf02cb73e387e86d1cff683102c91f45b3701054bcf97", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0c9f01d2779775803b448637ab281af195bd228ff344bb7e341e2917df604e80", "generation_seconds_batch": 22.855404519941658, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.641547679901123, "target_z": -0.059326093866274916, "ce": 6.308549880981445, "genuineness": 4.457139015197754, "identity": 0.5767459869384766, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/b4d2f9ad74156733ea2cf02cb73e387e86d1cff683102c91f45b3701054bcf97.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "b53afc2dd33390916df48124d1a038f2d14263854bfa555c648962bb0205c277", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "76c93de028b804b0847d54daeac00ca69430be9e0fc44b162f35189ee76390c1", "generation_seconds_batch": 23.311130142072216, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0580434799194336, "target_z": 0.45312507197659563, "ce": 6.09409761428833, "genuineness": 4.519899368286133, "identity": 0.7244486808776855, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/b53afc2dd33390916df48124d1a038f2d14263854bfa555c648962bb0205c277.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "b53b232eaa30b603429ce7829b7d9ea74ab7705a1e923460c67fdb9cb36a053b", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "20610fa4c4a833cb1bd31bfd5856b2b27dd04709b769d9a0f2b8829c77b82ff4", "generation_seconds_batch": 16.88415942597203, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1803598403930664, "target_z": 0.3457031176387573, "ce": 6.1020402908325195, "genuineness": 4.570107460021973, "identity": 0.7554084062576294, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/b53b232eaa30b603429ce7829b7d9ea74ab7705a1e923460c67fdb9cb36a053b.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "b61a3544e23e317c0ead00a8741492c27462cb1c1a400e9bc85b52d3899c3c33", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e7ba0bf399b9a5fe274ebb5e82a402419a88c0f83021c1e81942f97f6207f636", "generation_seconds_batch": 23.196196221979335, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9371817111968994, "target_z": 1.4375000261733075, "ce": 5.279973983764648, "genuineness": 4.607763767242432, "identity": 0.7306528091430664, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle prend.", "wer": 0.05263157894736842, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/b61a3544e23e317c0ead00a8741492c27462cb1c1a400e9bc85b52d3899c3c33.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "b67fe811e82c272509bee19951000c7c572a1ef85d6ebf072cbd6b132c719772", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "18577ba629be298eeb45ca77757da4fab7a6af00a2445da44f987d135e0f133b", "generation_seconds_batch": 21.199929359019734, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.466691017150879, "target_z": 0.09423835465794837, "ce": 6.062326908111572, "genuineness": 4.419482707977295, "identity": 0.7190060615539551, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/b67fe811e82c272509bee19951000c7c572a1ef85d6ebf072cbd6b132c719772.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "b6899b9c2edd951b1a43e0eac6dd6f284e3415ae35d34b3f731c7d5e44d58169", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "a7b13b9e46c29006ec8fdcb2fa95f3f93af73db5e89170d9747c41bba2b6ff02", "generation_seconds_batch": 17.743624301976524, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0580434799194336, "target_z": 0.45312507197659563, "ce": 6.181467056274414, "genuineness": 4.796044826507568, "identity": 0.6791750192642212, "hypothesis": "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/vn_STNC_reg_low/b6899b9c2edd951b1a43e0eac6dd6f284e3415ae35d34b3f731c7d5e44d58169.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "b69c2116ebab8fd3f57edcddb13dcc96f515f10fba5f2e4ff093c3e39af8f3af", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "de14ba1257dcc15e30d155df05d41d897879bedba42d9b922ab9f37ac48e9c39", "generation_seconds_batch": 15.981148398015648, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.052826166152954, "target_z": 1.3359375425316247, "ce": 5.974957466125488, "genuineness": 4.570107460021973, "identity": 0.31269943714141846, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/b69c2116ebab8fd3f57edcddb13dcc96f515f10fba5f2e4ff093c3e39af8f3af.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "b6a2d03ea8c2de0e0ad23e70c50da04a7cfd882dc558cab11b0c4b14c3a5118d", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2e4aaf3dadd9c2ddad3ad8e59249aeffe7b51c74533f76336e586293e78ec4bb", "generation_seconds_batch": 25.338029153994285, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6881012916564941, "target_z": 1.6562500392599613, "ce": 5.895530700683594, "genuineness": 4.833701133728027, "identity": 0.46908268332481384, "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/vn_STNC_reg_low/b6a2d03ea8c2de0e0ad23e70c50da04a7cfd882dc558cab11b0c4b14c3a5118d.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "b7115638cfd4e226d87b9d1f69d327922945ea18f8af097cea332d4bd83d4361", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "38323c88c48b820364580d0ba918646bba6fd40bed45eb3014d316074aa98ccc", "generation_seconds_batch": 22.682070187991485, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.205592632293701, "target_z": -0.5546875556182784, "ce": 5.935244083404541, "genuineness": 3.8546390533447266, "identity": 0.4193674921989441, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/b7115638cfd4e226d87b9d1f69d327922945ea18f8af097cea332d4bd83d4361.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "b7e45f46334fd537cf105c01171e0157f7516c9fd0324a623690968dc3097885", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "27312221ebb53736aaffb729f1a8f9bd489e548ab99c5071cd040fdfcc2ae0f0", "generation_seconds_batch": 16.62918314000126, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.035804271697998, "target_z": 0.4726562221908608, "ce": 5.990842819213867, "genuineness": 4.419482707977295, "identity": 0.7395615577697754, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/b7e45f46334fd537cf105c01171e0157f7516c9fd0324a623690968dc3097885.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "b88f5628256719dab4e0ab5ba6b300ca2dbd278fd9b0822aaad8a7c9fd5150ee", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5dcdfe501a5d9cc50f95b40593f73e1384b9b6694991066dad96d5132ab1163f", "generation_seconds_batch": 24.039401449030265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1506792306900024, "target_z": 1.25, "ce": 6.0146708488464355, "genuineness": 4.557555198669434, "identity": 0.6446192264556885, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/b88f5628256719dab4e0ab5ba6b300ca2dbd278fd9b0822aaad8a7c9fd5150ee.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "b8fa57689ec4e8ce787548c5eba03a020df16ef53ff93fe40ea5884542e2f78d", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5dcea2e7d0700fe7b3de18bd7d48b4f5c17c121f1616babf222caf1c7f4c443e", "generation_seconds_batch": 16.53627523209434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1684707403182983, "target_z": 1.234374954196712, "ce": 5.831989288330078, "genuineness": 4.7081804275512695, "identity": 0.29967111349105835, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/b8fa57689ec4e8ce787548c5eba03a020df16ef53ff93fe40ea5884542e2f78d.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "b9686246a3daa5eba83fe0956e81a67c5f9ff98f3a8e27a9df669d181f6c7fa6", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "5dd92a0448f082c384bdec6f2fe3606059cd69690bdcfbcca881fb0d5d94a3dd", "generation_seconds_batch": 16.954053697991185, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3641767501831055, "target_z": 1.0624999738266925, "ce": 5.919358730316162, "genuineness": 4.68307638168335, "identity": 0.6122076511383057, "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/vn_STNC_reg_low/b9686246a3daa5eba83fe0956e81a67c5f9ff98f3a8e27a9df669d181f6c7fa6.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "b983ea8c23547028bd43b1fd1c9395ae72b1bde25bfc9d7a527a95bb16e3c8a3", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "817d94e679bf517d2d40031e0ae33aada4c9103dcca8746e3a84be2f849f85d1", "generation_seconds_batch": 22.24303964106366, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3908638954162598, "target_z": 1.0390625098149904, "ce": 5.7128496170043945, "genuineness": 4.808597087860107, "identity": 0.7395308017730713, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/b983ea8c23547028bd43b1fd1c9395ae72b1bde25bfc9d7a527a95bb16e3c8a3.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "b9cc6a63bb9785c1ae2df85bf5bf48d4b7f538d8796df80e8fe1e04821c85f25", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "48449e6f7d5334720931b4e20eda02edea6eb1e7c526d886d30b1c65f14dd21b", "generation_seconds_batch": 18.00824969203677, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2574279308319092, "target_z": 1.1562500392599613, "ce": 5.331601142883301, "genuineness": 4.6203155517578125, "identity": 0.5730366706848145, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/b9cc6a63bb9785c1ae2df85bf5bf48d4b7f538d8796df80e8fe1e04821c85f25.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "b9df82e2dd36166737edd2abd09db9d15140c4d870bb32c9d8334b0a0fd9466c", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fa531ba33c8dec46790ca19f0dccea549ca12fe4bec400e2e1317071f9635375", "generation_seconds_batch": 20.804628188023344, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3997596502304077, "target_z": 1.0312499869133462, "ce": 5.168776512145996, "genuineness": 4.494794845581055, "identity": 0.5948733687400818, "hypothesis": "la puerta sigue abierta. Me 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/vn_STNC_reg_low/b9df82e2dd36166737edd2abd09db9d15140c4d870bb32c9d8334b0a0fd9466c.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "b9e98777107725e4d9915323a8819c4a3dbc42955c4891dce05b1f8d9997d343", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4f23e3f285fb8d4c58d77f8a634788bf9c92b1e3d3cc9a7eb04c7fcaca5f5d78", "generation_seconds_batch": 17.20774588396307, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8267545700073242, "target_z": 0.6562500392599613, "ce": 6.141753673553467, "genuineness": 4.6203155517578125, "identity": 0.2577304244041443, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/b9e98777107725e4d9915323a8819c4a3dbc42955c4891dce05b1f8d9997d343.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "ba2d104b6bbb382c581c4aa3efbe8a297ab701d8ab0a0807db1abe41ea3561e6", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "120d8de5a44fa49a91c74510fa3694ae76724d38f69ae59c88dc7a5432ec671f", "generation_seconds_batch": 17.027500704978593, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1558966636657715, "target_z": 0.36718742475174093, "ce": 6.141753673553467, "genuineness": 4.6705241203308105, "identity": 0.6042430400848389, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ba2d104b6bbb382c581c4aa3efbe8a297ab701d8ab0a0807db1abe41ea3561e6.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "ba5575dc1494e293245fddad6cc4446dfee1da5dee6a230ce84911a6d3045362", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4c9bb6e9ca7ee0e5b93d0b488bb646a221fd25e895b9f36e6c6a674f82e0f95c", "generation_seconds_batch": 21.496552930912003, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8267545700073242, "target_z": 0.6562500392599613, "ce": 6.133810997009277, "genuineness": 4.632867813110352, "identity": 0.28588739037513733, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ba5575dc1494e293245fddad6cc4446dfee1da5dee6a230ce84911a6d3045362.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "9fdd1afdbab900b49d59c59550ea4bcbd38f78489b360224651abd3a64689403", "generation_seconds_batch": 16.74950994201936, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8853464126586914, "target_z": -0.2734375687049322, "ce": 6.260893821716309, "genuineness": 4.6705241203308105, "identity": 0.8243523240089417, "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/vn_STNC_reg_low/ba70e7e208a6b15a9236ccd5a69e0e5fd403df2a937202d690effa2cdc45d9f2.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "ba7807d776675a3d04a3c92f8035ea39792b38ded215d3147cfd3c407c908075", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "454384441cc1806672cbef6e78c25fe5cbeed4453c8ab548ad27e60130ad5e87", "generation_seconds_batch": 21.431503625004552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.046923875808716, "target_z": 0.4628906470837282, "ce": 6.149696350097656, "genuineness": 4.65797233581543, "identity": 0.689285397529602, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ba7807d776675a3d04a3c92f8035ea39792b38ded215d3147cfd3c407c908075.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "ba9cd6983edad5484e84a26a059fdb1fe7e0e6ac2043c1d08152b7b053d23e46", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "02bb336b2115ccbfe4c157cab7567ec07da9253af6ac32cbb29702245a86f246", "generation_seconds_batch": 21.349195611081086, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6933186054229736, "target_z": 0.7734375687049322, "ce": 6.141753673553467, "genuineness": 4.7332844734191895, "identity": 0.6436421275138855, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ba9cd6983edad5484e84a26a059fdb1fe7e0e6ac2043c1d08152b7b053d23e46.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": "baaf7d37bd212c26121a518ea4294951e2031b2e530f77b6a9f5c84f92566e0f", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7fb2f153cb494e85d27d017a272435a702a51a647e6cd759e996b13f6ac0874a", "generation_seconds_batch": 16.80195873102639, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6532878875732422, "target_z": 0.8085937647224855, "ce": 5.967014789581299, "genuineness": 4.482243061065674, "identity": 0.5958214402198792, "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/vn_STNC_reg_low/baaf7d37bd212c26121a518ea4294951e2031b2e530f77b6a9f5c84f92566e0f.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "bb1c562cafe2afc4fa4e27d76b4db9a03cff7e20f54899779e539b8286edcdb4", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1e90421ce582cb111c4cbb8aaa03abc747f22ea4b38da7cf1595d46e8c925a9d", "generation_seconds_batch": 21.149435841012746, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9157118797302246, "target_z": 0.5781250196299806, "ce": 5.601652145385742, "genuineness": 4.080576419830322, "identity": 0.4535577595233917, "hypothesis": "The 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/vn_STNC_reg_low/bb1c562cafe2afc4fa4e27d76b4db9a03cff7e20f54899779e539b8286edcdb4.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "bb2437aaf05384c97569a623eaba1ca206e187bc127ec13569102b6b9421b390", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b5dfe335397103ef073243aa392ed7b965443e235a7b947514bc090bdda3c427", "generation_seconds_batch": 21.936192680033855, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.570659875869751, "target_z": 0.0029297353480777695, "ce": 6.070269584655762, "genuineness": 4.7332844734191895, "identity": 0.8271372318267822, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/bb2437aaf05384c97569a623eaba1ca206e187bc127ec13569102b6b9421b390.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "bb4c8a6bbde697c93101806153e44c1d825188e0a37f0b02856219d3f1e39a1b", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e189e6c197112ed1875cbac947d17e4aef8188bfa67b18e5f689a4580f4b9ef1", "generation_seconds_batch": 23.433784519904293, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4616873264312744, "target_z": 0.09863274829360505, "ce": 6.332377910614014, "genuineness": 4.582659721374512, "identity": 0.652351975440979, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/bb4c8a6bbde697c93101806153e44c1d825188e0a37f0b02856219d3f1e39a1b.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "bb4eb61a4d67f882ebdcd3922123f0dee1c29408eb68442170f0c2cbbd71ee63", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "78dd296d3ff686027eac8bdc03c002f00c203223b04e6efc64b4a8b48469dbbb", "generation_seconds_batch": 23.837759347981773, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.995773434638977, "target_z": 0.507812522901644, "ce": 6.062326908111572, "genuineness": 4.557555198669434, "identity": 0.6998496055603027, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/bb4eb61a4d67f882ebdcd3922123f0dee1c29408eb68442170f0c2cbbd71ee63.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "bb4f904bf500aa663f1b146a9c596ef7a8afb3d136066a9be3c6ab7895f09ba3", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e6bcd81f269966fa9d279be602d22179aaaed25667b82802d221510944ef69a1", "generation_seconds_batch": 23.110999403055757, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.470026969909668, "target_z": 0.09130861930987061, "ce": 6.205295085906982, "genuineness": 4.394378662109375, "identity": 0.8030210137367249, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/bb4f904bf500aa663f1b146a9c596ef7a8afb3d136066a9be3c6ab7895f09ba3.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "bb7e26730add34d78bd26872fdacc3cb6dd3d6f7bb9b8f29f1f52a3010522298", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fba6ee199b8db3cbee9ed6c919100637067e601a4986cc127a392f48033f5cf6", "generation_seconds_batch": 23.12531906401273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2574279308319092, "target_z": 1.1562500392599613, "ce": 5.839932441711426, "genuineness": 4.532451152801514, "identity": 0.5622798800468445, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/bb7e26730add34d78bd26872fdacc3cb6dd3d6f7bb9b8f29f1f52a3010522298.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "bba1c17dfc17026e5a681d58e29fa0d80a3df26f1d7347feddc0f97ac0b2dfb1", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "466de73d2f99d63851f2be1f09a0c76601702a996d9a97cc0faded0f195483fa", "generation_seconds_batch": 16.669484953046776, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9512948989868164, "target_z": 0.5468749280234044, "ce": 6.197352409362793, "genuineness": 4.494794845581055, "identity": 0.2337033450603485, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/bba1c17dfc17026e5a681d58e29fa0d80a3df26f1d7347feddc0f97ac0b2dfb1.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "bbce6c38287cb3c1b1068ac8133d52e9b8952c54be143d6961372fdad1e24726", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4ccd9ddbeb9bfcba5989e5c16a9c5dc06bebc316e2a632ce4dfe4db76d868540", "generation_seconds_batch": 21.678141291020438, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1328877210617065, "target_z": 1.265625045803288, "ce": 5.593709468841553, "genuineness": 4.607763767242432, "identity": 0.8526862859725952, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/bbce6c38287cb3c1b1068ac8133d52e9b8952c54be143d6961372fdad1e24726.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "bc7e336ead4b50f501eac90959c1364f90927dd976592cdf57255b46147c30cd", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a83482ee0961ad6d4aef70633bd9048e644991ddbf8b6b182c59233fab1a1226", "generation_seconds_batch": 22.855404519941658, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.724111318588257, "target_z": -0.1318359967988998, "ce": 6.324435234069824, "genuineness": 4.557555198669434, "identity": 0.6551072001457214, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/bc7e336ead4b50f501eac90959c1364f90927dd976592cdf57255b46147c30cd.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "bc820385e658fe6d578670641d316d8d9a9a53ae7ce4222db581967165f9e111", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b9165eeb97cd5ba07f4de70492789795d8c710c8f3e115dddcab20d67ee10f55", "generation_seconds_batch": 22.69683505699504, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.986877679824829, "target_z": 0.5156250458032882, "ce": 5.911416053771973, "genuineness": 4.394378662109375, "identity": 0.7677070498466492, "hypothesis": "De deur staat nog open. Ik 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/vn_STNC_reg_low/bc820385e658fe6d578670641d316d8d9a9a53ae7ce4222db581967165f9e111.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "bcb7d0ede40af88c37a4d40d7fb37512a29bd12c489d4b858cdbbf3b61d854ce", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "501b21a7fa17a4b104a785e1092e640ef09d14306724a2d9fa6f5a5dbf7fb9db", "generation_seconds_batch": 22.69683505699504, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.648840069770813, "target_z": 0.8124999738266925, "ce": 6.054384231567383, "genuineness": 4.394378662109375, "identity": 0.6690825819969177, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/bcb7d0ede40af88c37a4d40d7fb37512a29bd12c489d4b858cdbbf3b61d854ce.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "bd08dca2813300e63f5057ba36f71752212489b57b51bcbc8c7d08fe09461688", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "78970fff60ab803c2844e805e79eddfac7b5d2a192569b8e0bf460d7a7d8a0f2", "generation_seconds_batch": 23.03483499097638, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0973048210144043, "target_z": 1.2968750327166343, "ce": 6.046441555023193, "genuineness": 4.469690799713135, "identity": 0.6441294550895691, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/bd08dca2813300e63f5057ba36f71752212489b57b51bcbc8c7d08fe09461688.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "bd59616053a7ded5374fe31354f78cee984eec526ba987792adbed7d52ff67c0", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f824057b18aa2c83eb3ecc043236c0d1360d519e6de38be4c72b4f2970d4e401", "generation_seconds_batch": 22.638741133967414, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6529455184936523, "target_z": -0.0693360229722073, "ce": 6.356205940246582, "genuineness": 4.432034492492676, "identity": 0.6468154788017273, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/bd59616053a7ded5374fe31354f78cee984eec526ba987792adbed7d52ff67c0.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "bdd19f9707a44bc7b837cf168ca92eeb8e32f5d65d775eee5a64b015a7ca5bdd", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ebaffcae737de81b8f6c0588c4a6de458b37ca886b84892b8585ece3fa209df0", "generation_seconds_batch": 22.709262764896266, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9015988111495972, "target_z": 1.4687500130866538, "ce": 5.609594821929932, "genuineness": 4.268857479095459, "identity": 0.4743310213088989, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/bdd19f9707a44bc7b837cf168ca92eeb8e32f5d65d775eee5a64b015a7ca5bdd.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": "bdd7db6d0cf38ce3ad70ba3226988e8bb97dbe651c3c6593a1898ac02d333492", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ed142451b3eedaf89bed70344a12dcdf1c3315113bbcb981adf1c25d456d3da9", "generation_seconds_batch": 17.312285356922075, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8356503248214722, "target_z": 0.6484375163583171, "ce": 5.7684478759765625, "genuineness": 4.758388519287109, "identity": 0.6114903688430786, "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/vn_STNC_reg_low/bdd7db6d0cf38ce3ad70ba3226988e8bb97dbe651c3c6593a1898ac02d333492.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "be409285413da0bc92c32a3c20128bb5172f115f4bbfdaac5d3cdb05c2e973c6", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "71388571083486559ad4740d6b43b11435bd6be25cf106568565ceb2a01a224b", "generation_seconds_batch": 22.096640227013268, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8979203701019287, "target_z": 0.5937500654332687, "ce": 5.7684478759765625, "genuineness": 4.2249250411987305, "identity": 0.423490434885025, "hypothesis": "The 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/vn_STNC_reg_low/be409285413da0bc92c32a3c20128bb5172f115f4bbfdaac5d3cdb05c2e973c6.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "be7e73934b3d15d7bb1d138ed7d5d39dd80410318a354551ce169060a520de1a", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dccfeeb55ccc4e897f85d41fc26428863806d9b1411ba366d6617c6bbff0ca9a", "generation_seconds_batch": 16.34151714190375, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0343499183654785, "target_z": -0.4042969870544727, "ce": 6.324435234069824, "genuineness": 4.369274139404297, "identity": 0.49924659729003906, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/be7e73934b3d15d7bb1d138ed7d5d39dd80410318a354551ce169060a520de1a.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "be8ac8a54ded48a6811b3b83eced61908a51cacf14a3ee996be753a61224c304", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fbe14e563a716fcffe4ccbbe4737107030a6ef69c3f2c6d54c54252b1528c4e7", "generation_seconds_batch": 21.65205686201807, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1328877210617065, "target_z": 1.265625045803288, "ce": 5.6254801750183105, "genuineness": 4.337893962860107, "identity": 0.5588348507881165, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/be8ac8a54ded48a6811b3b83eced61908a51cacf14a3ee996be753a61224c304.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "beed68d3dbd45007f2ed6439b2648e2343f4607a2fb1fb9514fc0be11fc363d9", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "cf8bb90f736be3550e83239b0b1569e997fd28a05534895638936579a116db88", "generation_seconds_batch": 16.954053697991185, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3285938501358032, "target_z": 1.0937499607400387, "ce": 5.665193557739258, "genuineness": 4.607763767242432, "identity": 0.6302615404129028, "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/vn_STNC_reg_low/beed68d3dbd45007f2ed6439b2648e2343f4607a2fb1fb9514fc0be11fc363d9.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "c064193f6a3328ce16155a92da800f87a0625e14b1fb7fe5deae8306f40f5385", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0f2925029df016a44cb30fe79f903b07c48795f0d1a8e3ff81fd1abd4739d8f7", "generation_seconds_batch": 22.598050444037654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1239920854568481, "target_z": 1.273437464011702, "ce": 5.383228778839111, "genuineness": 4.570107460021973, "identity": 0.6626869440078735, "hypothesis": "J'ai trouvé la 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/vn_STNC_reg_low/c064193f6a3328ce16155a92da800f87a0625e14b1fb7fe5deae8306f40f5385.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "c09061441416562e890cf94487ceb24811f09f5bddd75573b96fcb56ef4d0895", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "363a71bec00424c2c35056bd96818e3124047746e6e4a271e1853ae4eba1b0be", "generation_seconds_batch": 21.896026425994933, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.251525640487671, "target_z": 0.28320314381206474, "ce": 6.372091293334961, "genuineness": 4.632867813110352, "identity": 0.3028722405433655, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c09061441416562e890cf94487ceb24811f09f5bddd75573b96fcb56ef4d0895.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "c0a2c6b6edb5fb83cf09d1854c7a405b0a1d90927193e216a93392704b4dfcf1", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4535a282014db42eb8e7ad81411eb43255af0a659ed546f3e10420ffd74d625a", "generation_seconds_batch": 18.510509132989682, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.236727714538574, "target_z": -0.5820313334274176, "ce": 5.990842819213867, "genuineness": 4.168440818786621, "identity": 0.38888391852378845, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c0a2c6b6edb5fb83cf09d1854c7a405b0a1d90927193e216a93392704b4dfcf1.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "c0ab77e46f3ca2f490aaf9ea685c2bf94311e6577c68ee331d21781d57af9e0c", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c9f8ca04ff1c294ce457c68db65cb69f1ca49c03981e7ae2a26b91ab0703948f", "generation_seconds_batch": 23.797784519963898, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0973048210144043, "target_z": 1.2968750327166343, "ce": 5.756534099578857, "genuineness": 4.808597087860107, "identity": 0.6714551448822021, "hypothesis": "He encontrado la cajita, estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c0ab77e46f3ca2f490aaf9ea685c2bf94311e6577c68ee331d21781d57af9e0c.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "c12038ff8808863f176f0ea1789ceb02f671d6891f710b6a617062af5195e5eb", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "adf36b2a5223559a96b3c75920db6896b3859172d4556e6c308598a562958175", "generation_seconds_batch": 16.524563557002693, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.577887773513794, "target_z": -0.003418024572757398, "ce": 5.974957466125488, "genuineness": 4.6203155517578125, "identity": 0.7951333522796631, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c12038ff8808863f176f0ea1789ceb02f671d6891f710b6a617062af5195e5eb.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "c1403ce6deb34e51b8889b592f91b28e7aaa962d640cd71d113e2432790046d0", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e1504388917bac21a635804e3cdde94f187135daaddedd9ab627b6c35f83bd8e", "generation_seconds_batch": 16.543650471023284, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7385668754577637, "target_z": -0.14453130725411015, "ce": 6.109982967376709, "genuineness": 4.381826400756836, "identity": 0.09612209349870682, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg, maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c1403ce6deb34e51b8889b592f91b28e7aaa962d640cd71d113e2432790046d0.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "c188314d21de65045c8720a1df7e28b8bc205c1d31bd407807c06134f4139f53", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "53e96b16ba3c4db4437571c52ad0102b5ffa11556e418b18210fe30ba65dcb7a", "generation_seconds_batch": 17.562141527072527, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5457696914672852, "target_z": 1.7812499869133462, "ce": 5.530168056488037, "genuineness": 4.959221839904785, "identity": 0.6916235685348511, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c188314d21de65045c8720a1df7e28b8bc205c1d31bd407807c06134f4139f53.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "c18fd1b8bb16866533a6bc33e1f5ada5726f134519e3163481b1bf5a18979a1a", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f2c8ee3c22066d0d6550eada9f8e324733b6a08bb8e84b30ffa1bac81130375b", "generation_seconds_batch": 17.84198633604683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.147000789642334, "target_z": 0.375000052346615, "ce": 6.173524379730225, "genuineness": 4.758388519287109, "identity": 0.796716570854187, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c18fd1b8bb16866533a6bc33e1f5ada5726f134519e3163481b1bf5a18979a1a.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "c1a63c24e1b81fd606cd7e42a97aa4a7c7707b2964f56c11422a97a08e35f7a5", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1196f6b56bbb35cf4e47b4073b8a6a5a8e34dc8a8af3e144dc00eadc66c4397f", "generation_seconds_batch": 16.61231451795902, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.004669189453125, "target_z": 0.5, "ce": 6.078212261199951, "genuineness": 4.582659721374512, "identity": 0.7082165479660034, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c1a63c24e1b81fd606cd7e42a97aa4a7c7707b2964f56c11422a97a08e35f7a5.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "c1e4f5f4a2bf8ca1375dfc31f00b13a564f2760ec1478a5597b419816e5457e6", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8615fa1f1b7439b76e98fac9fa9a1792c715b4689e59f040ecf1fd19670195c6", "generation_seconds_batch": 22.784538623993285, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.942399024963379, "target_z": 0.5546875556182784, "ce": 6.109982967376709, "genuineness": 4.369274139404297, "identity": 0.5956653356552124, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c1e4f5f4a2bf8ca1375dfc31f00b13a564f2760ec1478a5597b419816e5457e6.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "c1e9485d5048a9254a2ab8ac676ca86160b9fe117f3cd6426064ed8cf71b8ccf", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4d51b138a55c47d173211747909b65d6a36b1c3b029d6f0dd839fb732c4b1920", "generation_seconds_batch": 22.862096181022935, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3463852405548096, "target_z": 1.0781250196299805, "ce": 5.5539960861206055, "genuineness": 4.645420074462891, "identity": 0.5225052833557129, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c1e9485d5048a9254a2ab8ac676ca86160b9fe117f3cd6426064ed8cf71b8ccf.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "c1f418b459adbd68df9fdd29a8454f6a651efd51aa4a1eddcf7defeb4ee25faf", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "580cebe94a3030567563631ff96b3926b332d14d437a02093f035dedfda12650", "generation_seconds_batch": 23.37151700304821, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6221528053283691, "target_z": 0.8359375425316247, "ce": 5.665193557739258, "genuineness": 4.557555198669434, "identity": 0.8686666488647461, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c1f418b459adbd68df9fdd29a8454f6a651efd51aa4a1eddcf7defeb4ee25faf.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "c2659aad6240448c3d8a38f8682e828fd4e8c8ad04ac4c3496184e84454d4614", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9f5b5cb9271d89e4443a31c5e57f7838276dd96274ae2bad8b62f4613eb7a969", "generation_seconds_batch": 22.322207071003504, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.091402530670166, "target_z": 0.4238281372687379, "ce": 6.117925643920898, "genuineness": 4.68307638168335, "identity": 0.7548792362213135, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c2659aad6240448c3d8a38f8682e828fd4e8c8ad04ac4c3496184e84454d4614.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "c277e26e38f8b695eedcdb6e8c3648c102719ecb210a5364148dafaf1554ac7e", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "95cfc347d9d6edb8f8b08ea46b98718139bc04e39142322e5054744d47c70c27", "generation_seconds_batch": 17.013290524017066, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5821220874786377, "target_z": 0.8710937385491779, "ce": 6.09409761428833, "genuineness": 4.7081804275512695, "identity": 0.7640440464019775, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c277e26e38f8b695eedcdb6e8c3648c102719ecb210a5364148dafaf1554ac7e.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "c2973f2b286e8f9ca5277dc007df8a697fc73ac313073f93dc9328752df92726", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8e180c97c01c58891f1a5ae3e4f9effdb8fbff745a91af52a788cdba0d80c265", "generation_seconds_batch": 23.42693795298692, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3249154090881348, "target_z": 0.21875001308665376, "ce": 6.141753673553467, "genuineness": 4.469690799713135, "identity": 0.6602544784545898, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c2973f2b286e8f9ca5277dc007df8a697fc73ac313073f93dc9328752df92726.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "c2a708c6269e57c860bc0b49618c37a29d0da332c4e65bfb1be92b72b02c4459", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "90f0017d40d40fe91ddbe75507972576dafabb57c843c78ed135d6a75530379e", "generation_seconds_batch": 21.020450069918297, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5198519229888916, "target_z": 0.9257812941674564, "ce": 5.9431867599487305, "genuineness": 4.65797233581543, "identity": 0.7786886096000671, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c2a708c6269e57c860bc0b49618c37a29d0da332c4e65bfb1be92b72b02c4459.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": "c2be49731311cc39d9db8d7eaba98893fd5e6a38b1f52fbab5ca45a9f68c06ea", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7e70efbedea3bb883c4dcefa571395f2643c11a59f1bf636276f041255c23ddf", "generation_seconds_batch": 16.77627739601303, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.293010950088501, "target_z": 1.124999947653385, "ce": 5.462655067443848, "genuineness": 4.557555198669434, "identity": 0.6551625728607178, "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/vn_STNC_reg_low/c2be49731311cc39d9db8d7eaba98893fd5e6a38b1f52fbab5ca45a9f68c06ea.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "c330ad2c5458fb5a566375eac743a6527b311bec42414ff0425691f5abd47ab0", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d33b05013c1401ab579a3e4bdd6324c77ea9d5461f47f4df764596bd6431f73f", "generation_seconds_batch": 17.96961906296201, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0232300758361816, "target_z": -0.39453120256088015, "ce": 5.5817952156066895, "genuineness": 4.061748504638672, "identity": 0.7665714025497437, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c330ad2c5458fb5a566375eac743a6527b311bec42414ff0425691f5abd47ab0.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "96c77d127930c876666df6dd8ae18d08cb675dd438fe3972dd81a187b61d591e", "generation_seconds_batch": 17.956805411027744, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6140265464782715, "target_z": -0.035156196017553286, "ce": 6.117925643920898, "genuineness": 4.570107460021973, "identity": 0.5696868896484375, "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/vn_STNC_reg_low/c3977db0634952a7f9e4083e6135fbbc87199d34c7cb8c3338515d9ae38c18f4.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": "c39ea797da8abd5782cf364ed7107a4d47b116a0cd69f90537da0e4c7d5e0209", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e901b217d1c256089d4d71e2655a7a4fb3320a3755bfeb69ff828ff0f8cc5633", "generation_seconds_batch": 20.011200624052435, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2485322952270508, "target_z": 1.1640624574683753, "ce": 5.546053409576416, "genuineness": 4.7081804275512695, "identity": 0.8027173280715942, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai j'ai regardé dans le tiroir, puis près de la fenêtre. Prends-les quand tu partiras.", "wer": 0.041666666666666664, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c39ea797da8abd5782cf364ed7107a4d47b116a0cd69f90537da0e4c7d5e0209.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "c3a348279613cc8cd7d5d2c1a02c5ff268c8eef09f0e1505e867258fbb66dd5c", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "46b7cf1055ec0dce825ba41f474458cf32244cc21ae5e0e672a982dfff62b593", "generation_seconds_batch": 22.206867497996427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3730725049972534, "target_z": 1.0546874509250483, "ce": 5.959072113037109, "genuineness": 4.720732688903809, "identity": 0.7446678876876831, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c3a348279613cc8cd7d5d2c1a02c5ff268c8eef09f0e1505e867258fbb66dd5c.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "c448a8a4e422b11ece44a563a364df370649438b84a1dcde101f9588a0730ceb", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "04c563c2f0a92226f4232771da042c79bae401e880eaa094c624c823c9e18cdd", "generation_seconds_batch": 17.34863237803802, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1328877210617065, "target_z": 1.265625045803288, "ce": 5.172747611999512, "genuineness": 4.68307638168335, "identity": 0.6669530868530273, "hypothesis": "La puerta sigue abierta. he dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c448a8a4e422b11ece44a563a364df370649438b84a1dcde101f9588a0730ceb.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "023a71c7abfac4966a3ca500d1c37544298c855961914a2f110cb9b140e704a0", "generation_seconds_batch": 17.883747690008022, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.427985668182373, "target_z": -0.74999989530677, "ce": 6.229123115539551, "genuineness": 4.344170093536377, "identity": 0.4969743490219116, "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/vn_STNC_reg_low/c493619abd8899c9b3ce96f2a00a6bcdd900e874ef21436707b986365a8437ee.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "c4fa01b65473b66376de462f2e308b75eaecf257c69eb7a10046a3219b9698a7", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f7b8956a44f10e6036289d1bf1bffe75a8bdf9dd15b2b61be208f04ca0fb38d1", "generation_seconds_batch": 24.93546195095405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9112640619277954, "target_z": 0.5820312287341877, "ce": 5.847874641418457, "genuineness": 4.720732688903809, "identity": 0.67009437084198, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c4fa01b65473b66376de462f2e308b75eaecf257c69eb7a10046a3219b9698a7.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "c54b7186e736b806741ee081e517e1d38a9c6142594bc5251c49acb17685d615", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aa8acefa190e90d9b0c9f40f6728a65066d645ec1e0cad593cecbd1b566b62e9", "generation_seconds_batch": 21.099137362092733, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1736879348754883, "target_z": 0.35156258833491283, "ce": 5.967014789581299, "genuineness": 4.808597087860107, "identity": 0.6440801024436951, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c54b7186e736b806741ee081e517e1d38a9c6142594bc5251c49acb17685d615.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "c559e441db6729d84370482b3446e084d1942365936797f7d2cc1dba7dd3b895", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ce3631c27218b2d6d98487fe01b7602b88ba1b5204c5b3c7fd308d2b00125b4d", "generation_seconds_batch": 21.911886514048092, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1062005758285522, "target_z": 1.2890625098149904, "ce": 5.546053409576416, "genuineness": 4.582659721374512, "identity": 0.8282778263092041, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c559e441db6729d84370482b3446e084d1942365936797f7d2cc1dba7dd3b895.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": "c5842f55f254c766e0442b4acdc315796cb7a91a8b009ee2066da6e1d66c80b4", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b5eeaf1e970c2858a17801c121efa97d41b3385a3d77279519dc6f7bcd95a763", "generation_seconds_batch": 15.84039844200015, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.648840069770813, "target_z": 0.8124999738266925, "ce": 5.498397350311279, "genuineness": 4.570107460021973, "identity": 0.6980154514312744, "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/vn_STNC_reg_low/c5842f55f254c766e0442b4acdc315796cb7a91a8b009ee2066da6e1d66c80b4.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "c60f094fb00495303787580ee9e42394a023aff281c4b8bae9b2a872e0a31608", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1162c1f157421841dbfbff251b4fae40cb23248063f82bd038d0aede9a2d5508", "generation_seconds_batch": 17.27809655899182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6618411540985107, "target_z": -0.07714844118062136, "ce": 6.1020402908325195, "genuineness": 4.532451152801514, "identity": 0.5072001218795776, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c60f094fb00495303787580ee9e42394a023aff281c4b8bae9b2a872e0a31608.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "c6553c39fca2bf46035c89b7e43073e8b53bae00eef9735c859580c53023f40c", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6673d6e83117fb9ad5cc297ab9b81c58e63f4b7acbcd7794b5ba31929d39225c", "generation_seconds_batch": 24.498039223020896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9549732208251953, "target_z": 1.4218749803700195, "ce": 5.685050010681152, "genuineness": 4.532451152801514, "identity": 0.4382701814174652, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c6553c39fca2bf46035c89b7e43073e8b53bae00eef9735c859580c53023f40c.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "c67b7105e79dd8f9c40e038807b06c5bd4180777561b4d4db2be9bb44a476a72", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e4a71b6ec7a10da7170b7ce77e792fa3a9e8551f7c29b5264c6f2f95307b231e", "generation_seconds_batch": 16.149424509028904, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8045153617858887, "target_z": 0.6757811894742264, "ce": 5.8875885009765625, "genuineness": 4.532451152801514, "identity": 0.5410847067832947, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c67b7105e79dd8f9c40e038807b06c5bd4180777561b4d4db2be9bb44a476a72.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "c6db15777a7974839be7f328fb901656c1b74603db10068b283a760478e17ebb", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "88463e3d56af656ff95fd4a2d44d96789a6e9e6415646abf221625c5d9c4af01", "generation_seconds_batch": 23.21326634590514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.407200813293457, "target_z": 0.14648446415282868, "ce": 6.078212261199951, "genuineness": 4.871357440948486, "identity": 0.7770642042160034, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c6db15777a7974839be7f328fb901656c1b74603db10068b283a760478e17ebb.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "c7216998bbfc71cd4fd71893dfdbf88c0b6956142e067b256cc67c7266c04269", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ec1b7f1c9c3577aa807f3b6e008ffe7ed3e717560811085b9be00e63d2627936", "generation_seconds_batch": 17.450747145107016, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.547447681427002, "target_z": 0.023315391705532283, "ce": 6.252951145172119, "genuineness": 4.482243061065674, "identity": 0.6202774047851562, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c7216998bbfc71cd4fd71893dfdbf88c0b6956142e067b256cc67c7266c04269.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "c724cfd94bd887014f29318bb359dadec9cc3d90eaa557f884d026c4ade73d39", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e8316c6333830985f4900948564c2256bc034472d5d8d63ad4ce52f60ca5ee81", "generation_seconds_batch": 22.782622253056616, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4086554050445557, "target_z": 1.023437464011702, "ce": 5.502368927001953, "genuineness": 4.519899368286133, "identity": 0.5637796521186829, "hypothesis": "He encontrado la cajita. Esta junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c724cfd94bd887014f29318bb359dadec9cc3d90eaa557f884d026c4ade73d39.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "c743b21e60de8449185977a70be74393aa4ab3f6daf34469501c03e560bdd67b", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "527dd203713ce1f60dbca60eb6919509a4297ac8026638016989d83039c73452", "generation_seconds_batch": 21.9917241219664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8660159111022949, "target_z": 1.5, "ce": 5.700935363769531, "genuineness": 4.821148872375488, "identity": 0.8543286919593811, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c743b21e60de8449185977a70be74393aa4ab3f6daf34469501c03e560bdd67b.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "c762db9c25bd7175eeceb6ba44616b177eff1ca2fb56cb6644f3259b9d67db9b", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ea7e061e508427c8408a64998371d783b3781e4f28fcd18fd1425fb263f3d5f7", "generation_seconds_batch": 22.79148156696465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.212949275970459, "target_z": 1.1953125490749517, "ce": 5.792276382446289, "genuineness": 4.432034492492676, "identity": 0.666102409362793, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c762db9c25bd7175eeceb6ba44616b177eff1ca2fb56cb6644f3259b9d67db9b.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "c7799bbb6a9e5b3cb0b8630107d8fe0487fa2f9d69deee6d6a3e0b9a38ad6613", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2286220f4eca7cd9ba3d8c9dd80968fc45f8e1974dc16e661b2a72caace3a98a", "generation_seconds_batch": 18.305420147953555, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9468469619750977, "target_z": 0.5507812418208414, "ce": 6.062326908111572, "genuineness": 4.65797233581543, "identity": 0.6282504796981812, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c7799bbb6a9e5b3cb0b8630107d8fe0487fa2f9d69deee6d6a3e0b9a38ad6613.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "c80b848e3212afccd7a3d212beaa3e463dfd0155fd62d531c73a81372cea5c72", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "45638d53d5538257767a61146ebdf1e74da7244d5da1237e8715bd680d8bf8e3", "generation_seconds_batch": 16.50804863893427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.834195852279663, "target_z": -0.22851558819378634, "ce": 6.086154937744141, "genuineness": 4.6705241203308105, "identity": 0.48912614583969116, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c80b848e3212afccd7a3d212beaa3e463dfd0155fd62d531c73a81372cea5c72.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e51dd9de9e506721374fca32e8b5ae0ba4dce3ad9032e1e91784ab4ea749b444", "generation_seconds_batch": 17.95327451103367, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0276780128479004, "target_z": -0.3984375163583172, "ce": 5.974957466125488, "genuineness": 4.419482707977295, "identity": 0.6651150584220886, "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/vn_STNC_reg_low/c82131bfaf89850d394bf3a21117deb94436b65306054c3e1cbac0c65e2f913d.mp3" }, { "id": "heldout-en-3", "language": "en", "sentences": [ "Someone moved the little blue chair.", "It used to stand here... beside the tall bookshelf.", "We can put it back after lunch." ], "text": "Someone moved the little blue chair. It used to stand here... beside the tall bookshelf. We can put it back after lunch.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-3", "key": "c8907269f81c0a33f09ba3b9d9f76b6acb2a8250e6881dbc11786cea01719e9d", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "ca44b1dbfdbfb0ebb1ea9e39473565364951d05230ff33651a878e740522244b", "generation_seconds_batch": 15.47288943908643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2626452445983887, "target_z": 0.2734375687049322, "ce": 6.515059471130371, "genuineness": 4.6203155517578125, "identity": 0.3385593891143799, "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/vn_STNC_reg_low/c8907269f81c0a33f09ba3b9d9f76b6acb2a8250e6881dbc11786cea01719e9d.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "c96fc73cf3c296b91c27ee2cb9368892119465aa5059a637d3e77955586f9636", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "646e62fb0b63a1f055e535af459fbb3a653c75a869e6cd82becbe0173e1342bd", "generation_seconds_batch": 21.898873863043264, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.648840069770813, "target_z": 0.8124999738266925, "ce": 6.125868320465088, "genuineness": 4.8588056564331055, "identity": 0.8381065726280212, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c96fc73cf3c296b91c27ee2cb9368892119465aa5059a637d3e77955586f9636.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": "c9c11bafaf339c5fd8c53a8fb783c4f1f95c033eb4727f583e24e12fdc6e684b", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cd32530618add308ef304b98203816d94850f0a733f630ea3789a4eea1c235e3", "generation_seconds_batch": 16.77627739601303, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7422451972961426, "target_z": 0.7304687450925048, "ce": 5.474569320678711, "genuineness": 4.6956281661987305, "identity": 0.6135624051094055, "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/vn_STNC_reg_low/c9c11bafaf339c5fd8c53a8fb783c4f1f95c033eb4727f583e24e12fdc6e684b.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "c9d1cfc8749547c8995c5f760537c450561e1f895173d603d5b79ac6f1b7a623", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1dea30939b08f2f8d9461b9b1d2a7642628550fb6daad25b106a2a72a5e080d6", "generation_seconds_batch": 17.072704743011855, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9068162441253662, "target_z": 0.5859374378383947, "ce": 6.157639026641846, "genuineness": 4.758388519287109, "identity": 0.6691946983337402, "hypothesis": "Die Tür ist noch. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/c9d1cfc8749547c8995c5f760537c450561e1f895173d603d5b79ac6f1b7a623.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "ca05abc10d74bb44a265d7069a0df9cb6c0b11eede59484019c56e9bbb717b8a", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1c4dc1e21a2f9a6b482e89713317184248d7dff362e39e2ae257c465f3d7264f", "generation_seconds_batch": 23.426658542011864, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9549732208251953, "target_z": 1.4218749803700195, "ce": 5.597680568695068, "genuineness": 4.645420074462891, "identity": 0.715519368648529, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ca05abc10d74bb44a265d7069a0df9cb6c0b11eede59484019c56e9bbb717b8a.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "ca4673533ecc518f9c192eebbfdab9a15ad10e509c1831e884316456208af906", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8c94529355c69be2f6a15ac30ff908c9210817dab61d50fc85b31f7e8ef5e1a7", "generation_seconds_batch": 22.824129321961664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3285938501358032, "target_z": 1.0937499607400387, "ce": 5.911416053771973, "genuineness": 4.507347106933594, "identity": 0.5824738144874573, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ca4673533ecc518f9c192eebbfdab9a15ad10e509c1831e884316456208af906.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "ca48ffe77289508de58dae160608d092b935c413cc4fb6059c9cb3eae728741f", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4b88545dd2c0f3832b868c44b8039e1706dc88288678490a31dfb04e0943cb6f", "generation_seconds_batch": 21.715229570982046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.109194040298462, "target_z": 0.40820309146544975, "ce": 5.8875885009765625, "genuineness": 4.482243061065674, "identity": 0.6270670890808105, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ca48ffe77289508de58dae160608d092b935c413cc4fb6059c9cb3eae728741f.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "cadc14c42e2c643c26b5aa857b90d70bd965d13785725f572d3aff84ec2fb2bb", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c7251ecb709c2d4534ed49d0c588b8972e6ebb46dea02f7c86263382bbe6b65c", "generation_seconds_batch": 15.832757271942683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6221528053283691, "target_z": 0.8359375425316247, "ce": 6.316492557525635, "genuineness": 4.7081804275512695, "identity": 0.3372574746608734, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/cadc14c42e2c643c26b5aa857b90d70bd965d13785725f572d3aff84ec2fb2bb.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "cc4c3724d5e084f65e5b2707bc0c65bea7396c8045b48da79aecf9889faa6720", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e03c804d923f049c4778fe525ab248588ef6f58b444e315b54f25e5948780e21", "generation_seconds_batch": 16.845196036971174, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2493016719818115, "target_z": 0.2851563007107833, "ce": 6.006728172302246, "genuineness": 4.607763767242432, "identity": 0.7813921570777893, "hypothesis": "J'ai trouvé la la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/cc4c3724d5e084f65e5b2707bc0c65bea7396c8045b48da79aecf9889faa6720.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "ccc63e3e0e10398705012ebe8c9d32081f2cef656cdafd1d3c3d1059eb354768", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d64820039cd5508225420e6131efd4a8df7d8a34ef81f206ce70f270be7075bc", "generation_seconds_batch": 22.899351377040148, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1506792306900024, "target_z": 1.25, "ce": 5.847874641418457, "genuineness": 4.319066047668457, "identity": 0.6106929779052734, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ccc63e3e0e10398705012ebe8c9d32081f2cef656cdafd1d3c3d1059eb354768.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "cd72532be1b6c1341cf6f11ba64de8d5278825c01928d7fdb9cd31085fe8a518", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "538498ee1b0440cf75cd849ae7396916b22b180c8be76de02c83078769e81245", "generation_seconds_batch": 22.449539219029248, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.115865707397461, "target_z": 0.4023438301557542, "ce": 5.967014789581299, "genuineness": 4.595211505889893, "identity": 0.8008933663368225, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/cd72532be1b6c1341cf6f11ba64de8d5278825c01928d7fdb9cd31085fe8a518.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": "cdd4de3d8bc46f114a93362243243a54dac05b0ff52ffe10d3eaac6c0b9d573a", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The keys are on the kitchen table.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "694d0b9d360bb7024808a733d67e7d24a8fe3d54abc2cfb34a8ab592069bf321", "generation_seconds_batch": 17.197381179081276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.275219440460205, "target_z": 1.1406249934566732, "ce": 5.903473377227783, "genuineness": 4.632867813110352, "identity": 0.41304901242256165, "hypothesis": "The keys are on the kitchen table. I check the drawer. Then look beside the window. Take them with you when you leave.", "wer": 0.08695652173913043, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/cdd4de3d8bc46f114a93362243243a54dac05b0ff52ffe10d3eaac6c0b9d573a.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "cdd9d893d12b989af78383fd1b92a463b04722da7f8108b8e62833ebb92da6ca", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1a4b3e2b54f0724318c0b4e30e0d2da9ce3eed6a71eafcdff4bb6a63ceb1492e", "generation_seconds_batch": 21.393917390028946, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3019065856933594, "target_z": 1.117187529444971, "ce": 6.022613525390625, "genuineness": 4.570107460021973, "identity": 0.7025753259658813, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/cdd9d893d12b989af78383fd1b92a463b04722da7f8108b8e62833ebb92da6ca.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": "cdf3c81500caf4486e409e3874c54730e0c72ee1aa43aabe986e66e9ab78867f", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6259f553de81c04455e4e0ca5f14fd766594f70ebcdc9bfb42a365339e5f5e3a", "generation_seconds_batch": 15.845409853965975, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3374894857406616, "target_z": 1.0859375425316247, "ce": 5.331601142883301, "genuineness": 4.7081804275512695, "identity": 0.6949716210365295, "hypothesis": "J'ai écrit le numéro sur l'enveloppe. Lis-le bien et verifie les deux derniers chiffres. Appelle-moi quand tu ramènes un moment.", "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/vn_STNC_reg_low/cdf3c81500caf4486e409e3874c54730e0c72ee1aa43aabe986e66e9ab78867f.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "cdf8ee4d8cdfdb3935dc0e2522e2018389718895c2105ed9b12a8d50d83e9011", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5b8024ecaddaa2020fe1298a7799adec14a2b826600f086f37a3136615f85f73", "generation_seconds_batch": 23.03483499097638, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9460774660110474, "target_z": 1.4296875032716634, "ce": 5.839932441711426, "genuineness": 4.570107460021973, "identity": 0.5023627281188965, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/cdf8ee4d8cdfdb3935dc0e2522e2018389718895c2105ed9b12a8d50d83e9011.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "ce93c4db3cc3447ecfc4ee436e65ad4066c1a8bf676dbc3b65c4c004e07b6b86", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bb31686ffd95bdbeea8be5deb3dcd481c193f629edf1a7b3cbabec39f493a4e1", "generation_seconds_batch": 17.03882373892702, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7952771186828613, "target_z": -0.1943359706255923, "ce": 6.260893821716309, "genuineness": 4.319066047668457, "identity": 0.5641892552375793, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ce93c4db3cc3447ecfc4ee436e65ad4066c1a8bf676dbc3b65c4c004e07b6b86.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "ceb5fd4661d7b929eaa42a325c06e18ed87e6daa27c567ce1ce55abf12a311d5", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9ce966289dc3d07d19c4826248316735f169758c10cd95b5df47adb815beba23", "generation_seconds_batch": 22.393712619901635, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9690862894058228, "target_z": 0.5312499869133462, "ce": 6.268836498260498, "genuineness": 4.494794845581055, "identity": 0.42138877511024475, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ceb5fd4661d7b929eaa42a325c06e18ed87e6daa27c567ce1ce55abf12a311d5.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "cebba47626da81e4e84a165105a35820a338211a17cc0911d3838230ce3f9da2", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e1247c840df1bb4794e999668f73836c5d62f7b6b8505bae5f65037e3e68cab9", "generation_seconds_batch": 23.199840966961347, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0002212524414062, "target_z": 0.503906313797437, "ce": 6.078212261199951, "genuineness": 4.469690799713135, "identity": 0.5867471098899841, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/cebba47626da81e4e84a165105a35820a338211a17cc0911d3838230ce3f9da2.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "cee1bf3a0ea628808afd683274e20988a2b326c0ab427775075d0b47c5ce5ed7", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f9da60cb603a6e4a8e907b64c08a5bafdaee56590cfdc01aae2b756ceebba5d6", "generation_seconds_batch": 17.49798450199887, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.830859899520874, "target_z": -0.22558585284570856, "ce": 6.403861999511719, "genuineness": 4.519899368286133, "identity": 0.5924180746078491, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/cee1bf3a0ea628808afd683274e20988a2b326c0ab427775075d0b47c5ce5ed7.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": "ceee8ea28eb1eb66c919ab94e1e076b25c867244e1a4b95f0697a1a8aa3199b2", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ccb21417ee30b8f090c19b92086e153dc5a3ff2eb195bce1eedf00de1b5cbafe", "generation_seconds_batch": 20.78684511699248, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2040536403656006, "target_z": 1.2031249672833657, "ce": 5.784333229064941, "genuineness": 4.582659721374512, "identity": 0.7607075572013855, "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/vn_STNC_reg_low/ceee8ea28eb1eb66c919ab94e1e076b25c867244e1a4b95f0697a1a8aa3199b2.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "cf14c7ad1ca1b5165e0425f30f9831449e1c16e529e0ace8d3cbfee6655249a1", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ae3b4269f877b6385337fe201162efdca1a5e80e4d4dcc5a7748bee03fda6b68", "generation_seconds_batch": 21.904145609005354, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1736879348754883, "target_z": 0.35156258833491283, "ce": 6.41180419921875, "genuineness": 4.68307638168335, "identity": 0.5356271862983704, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/cf14c7ad1ca1b5165e0425f30f9831449e1c16e529e0ace8d3cbfee6655249a1.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "490991cc236ee4b0accfff230c13cf22ca55126a4ddf8929abfef6e4732051ad", "generation_seconds_batch": 15.681992769008502, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0410215854644775, "target_z": -0.4101562483641683, "ce": 6.292664527893066, "genuineness": 4.570107460021973, "identity": 0.5834097862243652, "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/vn_STNC_reg_low/cf68e6506caaaad7e7371e72de9d5417fef89211300487ff83acd2484eaab510.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "cf7dbc2c12113bf3b4a5b46d35ac163d9a8b4eb437eced00ddc7e2b6bf720fa8", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "56a619acd733d812476838c5de2fb858c0923e0128946ef244d5cf00696b280f", "generation_seconds_batch": 23.694935475010425, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.444238305091858, "target_z": 0.992187477098356, "ce": 5.990842819213867, "genuineness": 4.645420074462891, "identity": 0.5239848494529724, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/cf7dbc2c12113bf3b4a5b46d35ac163d9a8b4eb437eced00ddc7e2b6bf720fa8.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "d04be3bc10b076c8c1f4752f28492acd47fbb5890041c8cc9eb564a5f027191c", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dce7bb505ef954efc3b79ceb0fc3ca1fca7be4cc2e3724994be0694da3721a40", "generation_seconds_batch": 21.482714388053864, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6977665424346924, "target_z": 0.7695312549074952, "ce": 5.756534099578857, "genuineness": 4.6705241203308105, "identity": 0.6383566856384277, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d04be3bc10b076c8c1f4752f28492acd47fbb5890041c8cc9eb564a5f027191c.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": "d05c90c8ac9256e524a7b8fbc98685188fde09eb17a9de9c75a55a298bcc6eac", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9f1eb814831abcdf7fe9a6524ded20c2c27fb9cf6bbe62cb1357fc73437873a9", "generation_seconds_batch": 16.30163617595099, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8489940166473389, "target_z": 0.6367186796592361, "ce": 6.054384231567383, "genuineness": 4.607763767242432, "identity": 0.6906980276107788, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het naar de lunch terug.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d05c90c8ac9256e524a7b8fbc98685188fde09eb17a9de9c75a55a298bcc6eac.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "d0c36c0fad856bb861deab8950dd755e43be2f96287dc33210f90f2d57cb2d8e", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1bca20cc7ab5fb52d5b25c92bf3349258723e000cabaa12e670cf69945d846c3", "generation_seconds_batch": 17.199097343022004, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8430917263031006, "target_z": -0.2363282157886604, "ce": 6.332377910614014, "genuineness": 4.444586753845215, "identity": 0.5793954133987427, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d0c36c0fad856bb861deab8950dd755e43be2f96287dc33210f90f2d57cb2d8e.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "d0cefc5256a96d17d17b9d0b06701a6f14765850be9ad08db8c106a90113fd98", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4b044c7aa526ef0dcaa6eb0676a7eda2674e3e186c710dd7c889dd75d13044c4", "generation_seconds_batch": 22.45146721205674, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.202599048614502, "target_z": 0.3261719674244921, "ce": 6.1894097328186035, "genuineness": 4.632867813110352, "identity": 0.6576282978057861, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d0cefc5256a96d17d17b9d0b06701a6f14765850be9ad08db8c106a90113fd98.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "d119a07299663301f09543f1bce5742ada396c9ea75d4fa2e0c22dc6f005b122", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "049e8af3d65878d01e24f11d480dfbf4c27a50b0c4f8fc913b2514b230e519d9", "generation_seconds_batch": 16.459219877957366, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6206984519958496, "target_z": -0.04101566671370883, "ce": 6.141753673553467, "genuineness": 4.231201171875, "identity": 0.29845061898231506, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d119a07299663301f09543f1bce5742ada396c9ea75d4fa2e0c22dc6f005b122.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "43781e7b8aff1b39ca2786932f9ba1f2059a446af5ad0987267afbeaeeaf77cf", "generation_seconds_batch": 16.82580221700482, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.087724208831787, "target_z": -0.45117191507787713, "ce": 6.403861999511719, "genuineness": 4.231201171875, "identity": 0.5301499962806702, "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/vn_STNC_reg_low/d11caac31fa3732e2517803823f1b039213fb31e9438619fd83d8b232c894b7d.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "d1217e45a6945fc24bd1443b0ac8be9306bca7c12ae677245ea1198248a791b6", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "023ba7bedb8a0e940258280e433bb89ab248ecaa3829467577c91f7be4a88bed", "generation_seconds_batch": 22.775521537987515, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0261390209197998, "target_z": 1.3593750065433268, "ce": 5.677107334136963, "genuineness": 4.7458367347717285, "identity": 0.756855845451355, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d1217e45a6945fc24bd1443b0ac8be9306bca7c12ae677245ea1198248a791b6.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "d13ab45650913d539255d6c637fb33aeb953554f72f5a62b92a143f233300b29", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "bf5267f651301edf68704817e2311c8a3e933a704bf4a11815cecf8dacea67dd", "generation_seconds_batch": 16.01192609698046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7697019577026367, "target_z": -0.17187508506324936, "ce": 6.109982967376709, "genuineness": 4.093128204345703, "identity": 0.1546681672334671, "hypothesis": "The 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/vn_STNC_reg_low/d13ab45650913d539255d6c637fb33aeb953554f72f5a62b92a143f233300b29.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "d19019fc00b011fcd9ac82205186a8762f268eb8ee75d3bbd93cc381abb201d8", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0dc6f25a7dc84026629b46e2d13d6980ffc9f3eb31359e6541b38645b65c4fb6", "generation_seconds_batch": 22.142887281021103, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7708139419555664, "target_z": -0.1728516635126086, "ce": 6.252951145172119, "genuineness": 4.419482707977295, "identity": 0.610878050327301, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d19019fc00b011fcd9ac82205186a8762f268eb8ee75d3bbd93cc381abb201d8.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": "d19a082982b4e15ccdd9f30092e5f1822d3d460ac79fb316acd6ff086f88359c", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5bdc26880328bbe20e22944eae3b7bbd21e83801158587ee0f2ce09c7c5dc45e", "generation_seconds_batch": 20.164742170018144, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7022144794464111, "target_z": 0.7656249411100581, "ce": 5.673136234283447, "genuineness": 4.7458367347717285, "identity": 0.755391538143158, "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/vn_STNC_reg_low/d19a082982b4e15ccdd9f30092e5f1822d3d460ac79fb316acd6ff086f88359c.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "d1a57c610afa80efdffdd2bc80a487b8fa6eb43e7a1269a44c67b9ab9822ecba", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "90c17af222c5f548d8c965aaefd20e658bc7b8d05fb73a457dc2540e5712760b", "generation_seconds_batch": 21.688895749044605, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.387185573577881, "target_z": 0.1640624574683753, "ce": 5.927301406860352, "genuineness": 4.1998209953308105, "identity": 0.4245341718196869, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d1a57c610afa80efdffdd2bc80a487b8fa6eb43e7a1269a44c67b9ab9822ecba.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "d1aefd8ff184e9ad161fb6c06e6ddd8d065c10f9285921420d2f197cafc78bda", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1367d3b927a8ca7fa9ef1ef2eb8435e20ecd4d503e6119815512b592e944a2e0", "generation_seconds_batch": 17.49798450199887, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.706319808959961, "target_z": -0.11621095099561168, "ce": 6.395919322967529, "genuineness": 4.557555198669434, "identity": 0.5669355392456055, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d1aefd8ff184e9ad161fb6c06e6ddd8d065c10f9285921420d2f197cafc78bda.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "d205599ad7b5cd71829ac8c146f99d017a7e69353eb7e1cc3bd8e613eecefaf8", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "967d67d65afb03e9f3cc9dbddc870c954facb3a2dea97a5c4014ef2f107f46da", "generation_seconds_batch": 16.15646598499734, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8445460796356201, "target_z": 0.6406249934566731, "ce": 6.157639026641846, "genuineness": 4.7081804275512695, "identity": 0.693877100944519, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d205599ad7b5cd71829ac8c146f99d017a7e69353eb7e1cc3bd8e613eecefaf8.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "8596650ac09294ad6975d56c8a69e5132d44032b139db47544ea63b118b52b85", "generation_seconds_batch": 17.569804698927328, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.272310495376587, "target_z": -0.6132812156475339, "ce": 6.205295085906982, "genuineness": 4.432034492492676, "identity": 0.7523404955863953, "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/vn_STNC_reg_low/d2319f0ba4d4a13f099c61fd2524d2cd47bb09bca092815ed8a29e7f5b0c527c.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "d24d67cc238f5cef84fac687492c3b8845ec9e6c349feacb87f2e728be9db532", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2287f5ec29cb1cbb204397bda519c702377d12c19e214429774c262a86224332", "generation_seconds_batch": 16.183236254029907, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5906753540039062, "target_z": -0.014648467353928865, "ce": 6.125868320465088, "genuineness": 4.519899368286133, "identity": 0.5652841329574585, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand, tu es prêt ?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d24d67cc238f5cef84fac687492c3b8845ec9e6c349feacb87f2e728be9db532.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "d2ffde7308a69430607711495113f44203d790a36757267d20a94ada66abd0e9", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "03da0e5b9ce832fc1967ad76964f68b735c345675c4e29e4abda309f72823d24", "generation_seconds_batch": 21.020450069918297, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.221845030784607, "target_z": 1.1875000261733075, "ce": 5.708878040313721, "genuineness": 4.68307638168335, "identity": 0.6961647272109985, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d2ffde7308a69430607711495113f44203d790a36757267d20a94ada66abd0e9.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "d3185436651e6cca451074487cacc4bbd657c7b99589eaf896ea198caf25e5e0", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ad0798ff00b51c13f78d80ec603118cd6c51180ae1899255eae276ecb6141af7", "generation_seconds_batch": 16.873286899994127, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.782275915145874, "target_z": 0.6953125490749515, "ce": 5.804190158843994, "genuineness": 4.369274139404297, "identity": 0.7456505298614502, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d3185436651e6cca451074487cacc4bbd657c7b99589eaf896ea198caf25e5e0.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "d3c01f8d5fd4bf0ae4cabb5a690927e34dd9aa9f7348a798cbe480c0cb24d9fa", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "210ff87fad2a701753267be22f24c7d77d8c1066acedcda2815dd6fcefc1c952", "generation_seconds_batch": 21.936192680033855, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9335033893585205, "target_z": 0.5624999738266925, "ce": 6.149696350097656, "genuineness": 4.65797233581543, "identity": 0.8033257722854614, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d3c01f8d5fd4bf0ae4cabb5a690927e34dd9aa9f7348a798cbe480c0cb24d9fa.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "d3c148eb5f63c8391276cc73b5dc5d0f2922478f2e5d5264df737ce15af18485", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "afdf10d895d013bc6b29871f141d6696cc7c52803e6e0d25e60de1fb64f78f67", "generation_seconds_batch": 22.423571389983408, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.109194040298462, "target_z": 0.40820309146544975, "ce": 5.935244083404541, "genuineness": 4.582659721374512, "identity": 0.7503225803375244, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d3c148eb5f63c8391276cc73b5dc5d0f2922478f2e5d5264df737ce15af18485.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "d3ed8e2c0f8452c9b4b65b15e7cbaa9f9300faa473bf9242fa0c1149971ae11b", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5219730d9ac94648f59dae7e497d482cfdbe5447b0320b834e00b8e621e76f15", "generation_seconds_batch": 23.410492461989634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.657735824584961, "target_z": 0.8046874509250485, "ce": 6.038498878479004, "genuineness": 4.68307638168335, "identity": 0.6245365738868713, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d3ed8e2c0f8452c9b4b65b15e7cbaa9f9300faa473bf9242fa0c1149971ae11b.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "092430fe2b56107412b6ebd3f10207eb9a92dc21f7591712216f3d83300ca103", "generation_seconds_batch": 16.20624463201966, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.161113977432251, "target_z": -0.5156250458032882, "ce": 5.879645347595215, "genuineness": 4.293961524963379, "identity": 0.702583909034729, "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/vn_STNC_reg_low/d3fab0ead1fc9dcf899f59bddb0c9d4b54373ba1829b6dc49f77fc4027ce1740.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "d41ef2d911de1c89bc1dd3bb57efdbf793cc3588f008e264171776c946d89ffe", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9aef893ee47e2a6160eec94ceeea916c664c44f209eadeff004c4b6fe5c7f930", "generation_seconds_batch": 22.463476681965403, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.275219440460205, "target_z": 1.1406249934566732, "ce": 5.752562999725342, "genuineness": 4.6203155517578125, "identity": 0.7537758946418762, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d41ef2d911de1c89bc1dd3bb57efdbf793cc3588f008e264171776c946d89ffe.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "d44d2b3c0e97ca922f05c4b370d7ef50bebbe845807571f0ba06d69af83ca38c", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1b1d72c375590a6c5b00cc039ef044387ee13e7d6a4c07a51c38c18f9e78a7f5", "generation_seconds_batch": 22.719563339021988, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.417551040649414, "target_z": 1.015625045803288, "ce": 6.403861999511719, "genuineness": 4.7081804275512695, "identity": 0.2908661961555481, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d44d2b3c0e97ca922f05c4b370d7ef50bebbe845807571f0ba06d69af83ca38c.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "d4cf81a5b55fb82ebea6d76693e2407406c172045da49ff97bcd51fa33f55239", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "27d2e37bf5ef8dedcbecf5b5336ca6b6aaceb5ea6e4aedac749982fa1ed94cfd", "generation_seconds_batch": 23.509212131029926, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3160197734832764, "target_z": 0.2265624312950678, "ce": 5.982900142669678, "genuineness": 4.406930446624756, "identity": 0.4535132944583893, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d4cf81a5b55fb82ebea6d76693e2407406c172045da49ff97bcd51fa33f55239.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "d4fe29ef04e12a6e52e60e4aa24418d18cacda1adc02d3a4a9c7256facdde9f3", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b8bc79ca8d6baae27f4b265f4e284e26f667c3a112a418349dd28dfbb283042e", "generation_seconds_batch": 23.043923992081545, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8445460796356201, "target_z": 0.6406249934566731, "ce": 6.157639026641846, "genuineness": 4.607763767242432, "identity": 0.6434249877929688, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d4fe29ef04e12a6e52e60e4aa24418d18cacda1adc02d3a4a9c7256facdde9f3.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": "d5b0a8fc4b4ef760cf8ca010bc79a62a4deffd44831b91c653e605f9ad1f7ca7", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "710e89bebb221315df0083a515e88dea958a251daf24a7a9d7a7f150dddec9c1", "generation_seconds_batch": 17.312285356922075, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.488716959953308, "target_z": 0.9531249672833656, "ce": 5.736677646636963, "genuineness": 4.7332844734191895, "identity": 0.6402199864387512, "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/vn_STNC_reg_low/d5b0a8fc4b4ef760cf8ca010bc79a62a4deffd44831b91c653e605f9ad1f7ca7.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "d5d52faa8dd97d553652a9c4ef4dccab71c0ea3cc410837c103fae4a945a3683", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c471c2fae262ceba8aa16e158455bbfa27d7c4874bb39e73a15a2c15dbcd5757", "generation_seconds_batch": 16.669484953046776, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.417551040649414, "target_z": 1.015625045803288, "ce": 6.197352409362793, "genuineness": 4.645420074462891, "identity": 0.5297729969024658, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d5d52faa8dd97d553652a9c4ef4dccab71c0ea3cc410837c103fae4a945a3683.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "d5e3a859a2f4a31ff8504d2f886ab084dd05d96c1fd79689bd318b3133969d91", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "05b6150b4ede7da4ca162214d9f5e9e91fbf32d166ed4f5d937fe55fb9adcf92", "generation_seconds_batch": 22.879271298996173, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0091171264648438, "target_z": 0.496093686202563, "ce": 6.252951145172119, "genuineness": 4.808597087860107, "identity": 0.7219698429107666, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d5e3a859a2f4a31ff8504d2f886ab084dd05d96c1fd79689bd318b3133969d91.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "d66b4816ec7242c15bc44517a76e19873f0961a52a18f0bc6379b555b501613f", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "beaf5d843c9bcc6c8df7d073116f08c5d51574d253d22545ea364ebfbc12add0", "generation_seconds_batch": 22.986409360077232, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.648840069770813, "target_z": 0.8124999738266925, "ce": 5.855817794799805, "genuineness": 4.65797233581543, "identity": 0.8013200163841248, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d66b4816ec7242c15bc44517a76e19873f0961a52a18f0bc6379b555b501613f.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": "d685dc5a8ba1ae17f612f9f51aac4f6bfa997617f749e0772bf96f76a8b07ca2", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0cb8dadec3d09098640c9075900b80ad0328fcfcd68cc66a2da1f22e3a41915b", "generation_seconds_batch": 20.541158520034514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1684707403182983, "target_z": 1.234374954196712, "ce": 5.796247482299805, "genuineness": 4.720732688903809, "identity": 0.5028706789016724, "hypothesis": "De sleutels liggen op de keukentafel. Ik ik in de lach en daarna naast het raam, neem ze mee als je weg gaat.", "wer": 0.18181818181818182, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d685dc5a8ba1ae17f612f9f51aac4f6bfa997617f749e0772bf96f76a8b07ca2.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "d6b43bc1ad8ed2eabd9c389daba600b5095e9bf86bfc4a54e299028b59d6e0e0", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b1ab106ad64cc0e331a6258912a5055824b5dfd2795d464f43cbf0034ad6efd1", "generation_seconds_batch": 18.305420147953555, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5598827600479126, "target_z": 0.8906249934566731, "ce": 6.0146708488464355, "genuineness": 4.645420074462891, "identity": 0.7602899670600891, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d6b43bc1ad8ed2eabd9c389daba600b5095e9bf86bfc4a54e299028b59d6e0e0.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "d6f3a531bce8df0730513922fbfe8f58ddf637c64d92626e8ee30990956b267f", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d9075e27dd2bf67c1b0a504566c9519b05b406621930ba33d71775e9de1dd541", "generation_seconds_batch": 23.609800353995524, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3641767501831055, "target_z": 1.0624999738266925, "ce": 5.903473377227783, "genuineness": 4.3881025314331055, "identity": 0.5166304707527161, "hypothesis": "The door is still open. I I put your coat on the chair. Tell me when you are ready.", "wer": 0.05555555555555555, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d6f3a531bce8df0730513922fbfe8f58ddf637c64d92626e8ee30990956b267f.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "d70d0a778c1e9c09a8286551393ff5b9372ea12fec3ad906f4ea9550d04c189f", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "070323191b3978ba0981742c68aaf57819991983e75c04d07a51f901e5e70d12", "generation_seconds_batch": 22.716950829024427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9727646112442017, "target_z": 1.4062500392599613, "ce": 5.133034706115723, "genuineness": 4.582659721374512, "identity": 0.567970871925354, "hypothesis": "La puerta sigue abierta. he dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d70d0a778c1e9c09a8286551393ff5b9372ea12fec3ad906f4ea9550d04c189f.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "d71a4d77a02dc5ab50b1794b491c02e186af735fa8d057c1b3a33db65eae8edd", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1dc00b9aa72d315aa7b2ed8087a217b911d03afac11a7e4aae3f90db83620213", "generation_seconds_batch": 22.682953875046223, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7289016246795654, "target_z": 0.742187477098356, "ce": 6.0146708488464355, "genuineness": 4.557555198669434, "identity": 0.8359224200248718, "hypothesis": "La porte est encore ouverte j'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d71a4d77a02dc5ab50b1794b491c02e186af735fa8d057c1b3a33db65eae8edd.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "d74606a4ac6c2f55de5e91631d04d036332c408e13256916e7a2e55af6c7ccb2", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "108b93409a9b73f24a6a5632f0fdae8ff6055a7cfd2bfcfa92663629d028e28e", "generation_seconds_batch": 16.7476477350574, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7285590171813965, "target_z": -0.13574210120987684, "ce": 5.967014789581299, "genuineness": 4.6203155517578125, "identity": 0.6664813756942749, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d74606a4ac6c2f55de5e91631d04d036332c408e13256916e7a2e55af6c7ccb2.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "d74c9cdf327e39f3590f733265cf98488a8cdf918ac084894f8dd8d5e298b081", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "398d2b979d69503272736f7682a48dd24e580243626a4f3024057ff3ddd8d21d", "generation_seconds_batch": 16.767505232943222, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2937803268432617, "target_z": 0.24609379089579297, "ce": 6.117925643920898, "genuineness": 4.7834930419921875, "identity": 0.7443789839744568, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d74c9cdf327e39f3590f733265cf98488a8cdf918ac084894f8dd8d5e298b081.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "d7670352ee3110862d6b46d8f4fe12bce72b43605bdd7e6b3712f21a932e240a", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "33255f5963186f3c06a097a6bb3f23a5570c396395595ee4d32cbd533e8578b1", "generation_seconds_batch": 22.638741133967414, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.497270107269287, "target_z": 0.06738286607348878, "ce": 6.316492557525635, "genuineness": 4.6203155517578125, "identity": 0.7281697988510132, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d7670352ee3110862d6b46d8f4fe12bce72b43605bdd7e6b3712f21a932e240a.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "d76ba21aa89aab9f58e1c9e424b825ba2d549109b0734ac79d3494ecc19226e5", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e5217a8994d5e44128c4d46665e9a5573104cde8645ef2f9f8e963ac8cc038f7", "generation_seconds_batch": 16.496118741924874, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8037458658218384, "target_z": 1.5546874509250483, "ce": 5.752562999725342, "genuineness": 4.582659721374512, "identity": 0.588334321975708, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d76ba21aa89aab9f58e1c9e424b825ba2d549109b0734ac79d3494ecc19226e5.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "d77cc669440aa0107b8eb0217bfe4e969e2ca7298722960b22b1ddf3c549d82a", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e5cdbe29037cb6fd6b98804a6d7628b4c243d16e03b6bf750002b8bf4fea081c", "generation_seconds_batch": 17.052191909984685, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7111101150512695, "target_z": 0.757812522901644, "ce": 6.109982967376709, "genuineness": 4.507347106933594, "identity": 0.7960177063941956, "hypothesis": "De deur staat nog open. Ik heb ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d77cc669440aa0107b8eb0217bfe4e969e2ca7298722960b22b1ddf3c549d82a.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "d889311e4b7ba5797ed8269ecb5fa7c8551e925ab33f96c4143c7bc993005811", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "14fa11fedaa58315e5986fff5d56a92e21908233cc623b28221dff08495fcdd0", "generation_seconds_batch": 21.61453690403141, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3108023405075073, "target_z": 1.1093750065433268, "ce": 5.696964263916016, "genuineness": 4.6705241203308105, "identity": 0.6609110832214355, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d889311e4b7ba5797ed8269ecb5fa7c8551e925ab33f96c4143c7bc993005811.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "d94e5a1c214ce881079121ee12c3ac43ab09d5e0b9f9ad6ba4adfd4801e16776", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ea040cbaf47b376cb83edb8df79cc7ad67de08a8457934e49fcf6e90afe9ec6d", "generation_seconds_batch": 22.374279395095073, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5554349422454834, "target_z": 0.8945312025608801, "ce": 5.935244083404541, "genuineness": 4.645420074462891, "identity": 0.6443445682525635, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d94e5a1c214ce881079121ee12c3ac43ab09d5e0b9f9ad6ba4adfd4801e16776.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "d9cad0d457f07ce10a543debf298db7b61372e5a95fb868bb0b953be8097f971", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f8c2dfefeb5763bd118c70445b1b749ba98548999b5654ff9e1a7f8b04718792", "generation_seconds_batch": 16.715432905009948, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.671079397201538, "target_z": 0.7929687189191974, "ce": 6.0305562019348145, "genuineness": 4.243753433227539, "identity": 0.4400414228439331, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/d9cad0d457f07ce10a543debf298db7b61372e5a95fb868bb0b953be8097f971.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "da0144536b3c65f03ea8789beb342327d55ffc425029c045fc365c1b5e15eec4", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "a96f053d624b864d66932c92bd85ba65f80d64aeeb55d91dc6500928c073d47a", "generation_seconds_batch": 22.29878579999786, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8934725522994995, "target_z": 0.5976562745374758, "ce": 6.157639026641846, "genuineness": 4.7458367347717285, "identity": 0.6297920346260071, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/da0144536b3c65f03ea8789beb342327d55ffc425029c045fc365c1b5e15eec4.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "da11d0a72a832b8ad1009d612a6cb5554afc2177f691ad653bd6b0999b6825c1", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f95a7aa6a28200551d0debe9117ad8b09d4e32ce95d00940936f634eb7473a35", "generation_seconds_batch": 23.77284389093984, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9913256168365479, "target_z": 0.511718732005851, "ce": 6.292664527893066, "genuineness": 4.3504462242126465, "identity": 0.35479483008384705, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/da11d0a72a832b8ad1009d612a6cb5554afc2177f691ad653bd6b0999b6825c1.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "daf4fe1a5217f3a3d1338ea338b85cd9e117a9a7bee822907f35c630baab4bb8", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "51b551a75ccfeab54a6dced0a01df90e042fcf357c340d37a0fde92ff85ff6b4", "generation_seconds_batch": 17.41752178000752, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.178135871887207, "target_z": 0.3476562745374758, "ce": 6.141753673553467, "genuineness": 4.406930446624756, "identity": 0.5388355255126953, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/daf4fe1a5217f3a3d1338ea338b85cd9e117a9a7bee822907f35c630baab4bb8.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "7227d532e32bbd43c90d357dacc194bff4cbf5ce33dcee37b0003d280cf11bff", "generation_seconds_batch": 18.000636689015664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7952771186828613, "target_z": -0.1943359706255923, "ce": 6.4912309646606445, "genuineness": 4.369274139404297, "identity": 0.39135417342185974, "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/vn_STNC_reg_low/db192910e9ae67de251fb2a990940eab881f66f627c78b50b638ce61f5a63679.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": "db1ad24f96c4c1aa73b6491e3b5524342e27e76cdbae94986d7c5461498cb48c", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8312a14f17bbb8dd90650c999ebe95e9860373d54fa61aaeb12666f06ea428d9", "generation_seconds_batch": 20.705067010014318, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5065083503723145, "target_z": 0.9375000261733075, "ce": 5.363371849060059, "genuineness": 4.7081804275512695, "identity": 0.8381173014640808, "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/vn_STNC_reg_low/db1ad24f96c4c1aa73b6491e3b5524342e27e76cdbae94986d7c5461498cb48c.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "db36f9362b9e09f0a070a32ded9c53de72c516cc5ccf3ec6ec91e307f7c2107c", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "746b05429a45822a89f8aaa9d54e27bdfd64071a06ceda934fde141fd82d3171", "generation_seconds_batch": 23.05403789097909, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8178589344024658, "target_z": 0.6640624574683753, "ce": 6.046441555023193, "genuineness": 4.7834930419921875, "identity": 0.8531824350357056, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/db36f9362b9e09f0a070a32ded9c53de72c516cc5ccf3ec6ec91e307f7c2107c.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "db53b295ea9ceab21555a9261242e4ad2fd831b28a6cf4fd0f2c850b5e94b325", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6c8ce7a76f4524418561f6386d6057bc5fb136b18d815707871090d04f2f91ea", "generation_seconds_batch": 26.791728196083568, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9690862894058228, "target_z": 0.5312499869133462, "ce": 6.070269584655762, "genuineness": 4.883909225463867, "identity": 0.7992039918899536, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/db53b295ea9ceab21555a9261242e4ad2fd831b28a6cf4fd0f2c850b5e94b325.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "db61ca66e6255164bca2684590cfac37050485afac8117ac394a269297498150", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bd1ed6cd862f9ea2d495f88865ece57d4451ff483c518f8639cbfdbe58366313", "generation_seconds_batch": 17.040524665964767, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3193554878234863, "target_z": 0.22363290533345004, "ce": 5.784333229064941, "genuineness": 4.325342178344727, "identity": 0.6018762588500977, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/db61ca66e6255164bca2684590cfac37050485afac8117ac394a269297498150.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "db96acc3d8b5984bf8a2c091162662309f4bb9c488be2f555fe243de7a1c2230", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cb9c820cabeb3f553347df14e49eddaa5c1f326ed6185d8dc8fea2972386ad05", "generation_seconds_batch": 21.313703596009873, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1150963306427002, "target_z": 1.2812499869133462, "ce": 5.387199878692627, "genuineness": 4.65797233581543, "identity": 0.5349263548851013, "hypothesis": "La puerta sigue abierta. He dejado tu. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.26666666666666666, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/db96acc3d8b5984bf8a2c091162662309f4bb9c488be2f555fe243de7a1c2230.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "db9f1e30c7bf6f4583ad8f46ec31216a6608e51933a287ff5181b0573087f5ae", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c1eca42c3d82f1209f0ae26e6bc680e5568d1c60a0ef0e76156e68cf2a45eefd", "generation_seconds_batch": 17.072704743011855, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3805136680603027, "target_z": 0.16992192816453086, "ce": 6.23706579208374, "genuineness": 4.720732688903809, "identity": 0.6942100524902344, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/db9f1e30c7bf6f4583ad8f46ec31216a6608e51933a287ff5181b0573087f5ae.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "db9fbafde687c1527e850d986911ff6a4f8627d67b15458249078f225b7bda30", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2366556886527145c53b5382a9276ab9c6703ae9bdcebe01120f360bbb51eb59", "generation_seconds_batch": 22.423571389983408, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3527145385742188, "target_z": 0.1943359706255923, "ce": 6.141753673553467, "genuineness": 4.7834930419921875, "identity": 0.8649629354476929, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/db9fbafde687c1527e850d986911ff6a4f8627d67b15458249078f225b7bda30.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "dbfafaa37603477b8f7b6fa349af508b620134e2670dba2d90c742f4bf8593f1", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "902834db58249d50f198ac9fceede53712cf10ac92ad8ce39b02069ba40054f6", "generation_seconds_batch": 16.41353548504412, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9335033893585205, "target_z": 0.5624999738266925, "ce": 6.2767791748046875, "genuineness": 4.68307638168335, "identity": 0.48731306195259094, "hypothesis": "I found the little box. It was the side, the old window. We can look inside together.", "wer": 0.125, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/dbfafaa37603477b8f7b6fa349af508b620134e2670dba2d90c742f4bf8593f1.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "dc1018996f7230a3ae1740769fc5784dd2f9f0803a1226dbf1b602459ddda210", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8197dc0373d5de81a48499b2470f03f5601899fb4e3ce24c4e4ee7ebb7e69d12", "generation_seconds_batch": 17.54738755291328, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.507833957672119, "target_z": 0.05810537080457585, "ce": 5.967014789581299, "genuineness": 4.6956281661987305, "identity": 0.6758989095687866, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/dc1018996f7230a3ae1740769fc5784dd2f9f0803a1226dbf1b602459ddda210.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "dc3532fa2563a451287f3f395be00fa4e98c6bb4cf32d7aa9eb7a7423b8de5e1", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "515b3ae1a801fe10594253468534a76a98c5327b3b550cf30512a45aec54b6bb", "generation_seconds_batch": 16.972046782029793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8356503248214722, "target_z": 0.6484375163583171, "ce": 5.967014789581299, "genuineness": 4.582659721374512, "identity": 0.7296229600906372, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/dc3532fa2563a451287f3f395be00fa4e98c6bb4cf32d7aa9eb7a7423b8de5e1.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "dc62d9161321072acc0322b148094291294dd00759cec75bec8e3334af13755b", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "a21e1d63a9f4c0682527c1ea1fe007b04a691fa3565399cfd9b65f4998d1d007", "generation_seconds_batch": 21.430985065060668, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1247615814208984, "target_z": 0.39453120256088015, "ce": 6.205295085906982, "genuineness": 4.375550270080566, "identity": 0.5901147127151489, "hypothesis": "The 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/vn_STNC_reg_low/dc62d9161321072acc0322b148094291294dd00759cec75bec8e3334af13755b.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "dc8617a30aaeb4bfdb27ef10b34eb59cb8a57d22a9d61b7d61142af7c889abd9", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e1b1e26e8eaff02e47075e34606196a63a747a1d13ba0eb0758e07e13e294b11", "generation_seconds_batch": 23.868397870915942, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2396365404129028, "target_z": 1.1718749803700195, "ce": 5.649308204650879, "genuineness": 4.68307638168335, "identity": 0.6095236539840698, "hypothesis": "He encontrado la cajita, estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/dc8617a30aaeb4bfdb27ef10b34eb59cb8a57d22a9d61b7d61142af7c889abd9.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": "dd046ee5a3e0c69f2ab4fb3a38ce427c462e7ebe413e0ff5363ae3f6c5cd3356", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8aa4d0dcf5633ec82537cc46a2bdb4bde8617eb1d3ef0b1a615bc23e8188c3df", "generation_seconds_batch": 20.49811221694108, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.444238305091858, "target_z": 0.992187477098356, "ce": 5.744620323181152, "genuineness": 4.413206577301025, "identity": 0.8367521166801453, "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/vn_STNC_reg_low/dd046ee5a3e0c69f2ab4fb3a38ce427c462e7ebe413e0ff5363ae3f6c5cd3356.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": "de06da1fb7d673b2a4f19b772977cbc5411b064f812c24026b18cb70d473ac51", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "64b6dafe44b18ee7c2c52e97eb5f5f4410549a976f73513291324459e54ed5dd", "generation_seconds_batch": 20.173583226976916, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.088409185409546, "target_z": 1.3046874509250483, "ce": 5.095306873321533, "genuineness": 4.557555198669434, "identity": 0.7508048415184021, "hypothesis": "Ton message est arrivé après mon départ, j' je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/de06da1fb7d673b2a4f19b772977cbc5411b064f812c24026b18cb70d473ac51.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "de62ff8c560ff203db579fc83bdd014d65b8d64d6e7b90e886baa75900bfebd8", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f31a2b9ee478e15a2c069161ce66e72f2e6c4b47bafc00f1e2a60adab9f224cc", "generation_seconds_batch": 21.852705170982517, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.448686122894287, "target_z": 0.988281267994149, "ce": 5.808161735534668, "genuineness": 4.65797233581543, "identity": 0.7495836019515991, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/de62ff8c560ff203db579fc83bdd014d65b8d64d6e7b90e886baa75900bfebd8.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "defcb3df6409ef3dda6487f4b9b3e0f7d237649f8b1f3570b27b548c7f4de6c6", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9a4d7b7557b1410ce62ee097d79e63b73319388fabfced8d07a2b476643c700d", "generation_seconds_batch": 17.43679237004835, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1247615814208984, "target_z": 0.39453120256088015, "ce": 6.38003396987915, "genuineness": 4.7458367347717285, "identity": 0.31192636489868164, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/defcb3df6409ef3dda6487f4b9b3e0f7d237649f8b1f3570b27b548c7f4de6c6.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "df0a5548959f3c9f51d55352fdb55ca75a8ef8854ee8f044edb32676cfc1e5e3", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d32cb170f356c344a9ab32a8855a9029716aba64113453e97762afb65018bbe6", "generation_seconds_batch": 21.11886638798751, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.466477632522583, "target_z": 0.9726562221908608, "ce": 5.40705680847168, "genuineness": 4.632867813110352, "identity": 0.40474623441696167, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/df0a5548959f3c9f51d55352fdb55ca75a8ef8854ee8f044edb32676cfc1e5e3.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "e04608d8e73fd359ea4d018392403bd270c4e1c392d44fac8f3264ce4dfb6550", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "764b570629e61fd9577c4a3c763d74879e55e86c0f6e1512acc36af458c55212", "generation_seconds_batch": 16.53948803897947, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2604212760925293, "target_z": 0.2753907256036507, "ce": 6.157639026641846, "genuineness": 4.507347106933594, "identity": 0.6597051024436951, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/e04608d8e73fd359ea4d018392403bd270c4e1c392d44fac8f3264ce4dfb6550.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "e0774a64168ce7546ef1395e6f7f086c040665421b2ed97cedc33179d48189b1", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dd71985781edebf4dd71ca79aa312b5e057af889eae242150f9a2b66836c611c", "generation_seconds_batch": 16.390651933965273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5353550910949707, "target_z": 0.033935472955854216, "ce": 6.117925643920898, "genuineness": 4.796044826507568, "identity": 0.6546536684036255, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/e0774a64168ce7546ef1395e6f7f086c040665421b2ed97cedc33179d48189b1.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "e144cf2a1546ec6c627f399dbdc9ae556a8e709252bb880cb68672e76070677f", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ff0e2c3ee23025e33402b57451266a7613297181b889ee49efc61ce0494fdca0", "generation_seconds_batch": 23.400134368101135, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7644845247268677, "target_z": 0.7109374901850097, "ce": 6.141753673553467, "genuineness": 4.582659721374512, "identity": 0.6714766621589661, "hypothesis": "Ich hab die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/e144cf2a1546ec6c627f399dbdc9ae556a8e709252bb880cb68672e76070677f.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": "e1eaae9efeceffd3fe1c34672f1616865dfedd0da511a483a8ce4f1c0b2a3365", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0ada05312cd7a0bb148dfd494860b0e4328772ee59a3759c83265cf7acc7e7d9", "generation_seconds_batch": 20.724322946043685, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2574279308319092, "target_z": 1.1562500392599613, "ce": 5.677107334136963, "genuineness": 4.65797233581543, "identity": 0.7198877930641174, "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/vn_STNC_reg_low/e1eaae9efeceffd3fe1c34672f1616865dfedd0da511a483a8ce4f1c0b2a3365.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "e232bfbae1bd1579bf48d546c8ea0f43255d00aef9ab7ae278a45c5ad9e4923b", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e6ffe0de03ab68cc95c4b77aa6a6b5130a54e2c05bba9850f04fe3f1ed8c5eb8", "generation_seconds_batch": 17.099203041987494, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.636265993118286, "target_z": -0.054687555618278434, "ce": 6.109982967376709, "genuineness": 4.632867813110352, "identity": 0.6890456676483154, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/e232bfbae1bd1579bf48d546c8ea0f43255d00aef9ab7ae278a45c5ad9e4923b.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "e2b19a699163ab45ce1640210b7097e933af7c47652568d1842dfc904b0ad6b5", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bcbb7ebc4391d496f56e7fa6613b0fc0550709158657eaf49df4863c7f94493a", "generation_seconds_batch": 16.965796970995143, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5242998600006104, "target_z": 0.9218749803700194, "ce": 6.046441555023193, "genuineness": 4.595211505889893, "identity": 0.2026253342628479, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/e2b19a699163ab45ce1640210b7097e933af7c47652568d1842dfc904b0ad6b5.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "e2cdf3344d20d3c401b237393f54c37334043f52da65f7c8d447dae2bdb48c21", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7dcabb3931c6b034b0092d0a293001cb29a6bb9ba19f73e72bf8f7d9c146c746", "generation_seconds_batch": 16.296801679069176, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9431686401367188, "target_z": -0.3242188105257736, "ce": 6.332377910614014, "genuineness": 4.268857479095459, "identity": 0.571776270866394, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/e2cdf3344d20d3c401b237393f54c37334043f52da65f7c8d447dae2bdb48c21.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "e30047b01e9b64d1e22a7c4cffec1463bbb0a74e6af7bd3c6a36288ad8b9a939", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fb4b92a9bedf9fbcb2b70385c5dcfa0bc53f66caad2c5b2dd1322f7372f2d6e9", "generation_seconds_batch": 21.523821343900636, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.48392653465271, "target_z": 0.07910159807933988, "ce": 6.086154937744141, "genuineness": 4.68307638168335, "identity": 0.6657280325889587, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/e30047b01e9b64d1e22a7c4cffec1463bbb0a74e6af7bd3c6a36288ad8b9a939.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "e331f6391758aabb9f1508f7f6d46742d6d32b283fd72ce7272902f1acbb01f5", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f7301e085dd9a0fd780130a9a3b0f4ecc748ac36fb62ceb38704be3e802ae022", "generation_seconds_batch": 22.923548455932178, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9023683071136475, "target_z": 0.5898437516358317, "ce": 5.855817794799805, "genuineness": 4.005263805389404, "identity": 0.4851158857345581, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/e331f6391758aabb9f1508f7f6d46742d6d32b283fd72ce7272902f1acbb01f5.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "e334f93c6a94ea34fadd50bea563bd8921e9e58fc28206733c942063151b9d57", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1397a292a56d6624f32d47f4ad1e414833f6f3c4589fb0132f6cbe19f75cdbd2", "generation_seconds_batch": 21.541966571006924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9994518756866455, "target_z": 1.382812470555029, "ce": 5.700935363769531, "genuineness": 4.457139015197754, "identity": 0.3779452443122864, "hypothesis": "I found the little box. It it was beside the old window. We can look inside together.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/e334f93c6a94ea34fadd50bea563bd8921e9e58fc28206733c942063151b9d57.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "e398e4d4f2e8e89cc5d0a45379c4df96205b9dd703ab186b03066a747e5cb7eb", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e68a445234ef4b054fd6820cb35d5c6afe987a36dfea52112336320debf24bd1", "generation_seconds_batch": 22.561512141954154, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5198519229888916, "target_z": 0.9257812941674564, "ce": 5.903473377227783, "genuineness": 4.6705241203308105, "identity": 0.5729027390480042, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/e398e4d4f2e8e89cc5d0a45379c4df96205b9dd703ab186b03066a747e5cb7eb.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "e39b5e1a030352c211443565726ab7a269fa098d1be34ca1f1740879c68272d4", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4ff1d617003428b06bcad0a78073a603b1045208ed0171ad5065c45e8158022e", "generation_seconds_batch": 22.463476681965403, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5687785148620605, "target_z": 0.8828124705550291, "ce": 6.070269584655762, "genuineness": 4.7081804275512695, "identity": 0.5728437900543213, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/e39b5e1a030352c211443565726ab7a269fa098d1be34ca1f1740879c68272d4.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "e3dcb42781df3fcfdf9647a758cd068d1d63b4fd7e5028c17829a4f2359dbd3d", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "37c6c1b458e96900712e8e011dcbcb24036fbd8a63b5b1f5f7826244f635f9fb", "generation_seconds_batch": 16.858227350981906, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.515061616897583, "target_z": 0.05175782027020066, "ce": 6.260893821716309, "genuineness": 4.6705241203308105, "identity": 0.7042734622955322, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/e3dcb42781df3fcfdf9647a758cd068d1d63b4fd7e5028c17829a4f2359dbd3d.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "e3e614becf21526dad6997b962ac785a609fa1aba4e6d0ac8d0b5e6014f13a82", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "634e1be7be686445610f00d673c13c2ddcf94ba5ee0e9ab68892b8652f27504c", "generation_seconds_batch": 22.297051931964234, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3196980953216553, "target_z": 1.1015624836416829, "ce": 5.959072113037109, "genuineness": 4.607763767242432, "identity": 0.2412288784980774, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/e3e614becf21526dad6997b962ac785a609fa1aba4e6d0ac8d0b5e6014f13a82.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": "e3f65ebcabdd0e89ae51713adaf1a76277b1ba48dbca3ff234859abad473d536", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ffd9d5819ee1cc12fbf25ffb3e013e88fa50d813ebce502f2d87f64fcfe01fff", "generation_seconds_batch": 16.01888156903442, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7733802795410156, "target_z": 0.7031249672833656, "ce": 5.831989288330078, "genuineness": 4.7332844734191895, "identity": 0.6310499906539917, "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/vn_STNC_reg_low/e3f65ebcabdd0e89ae51713adaf1a76277b1ba48dbca3ff234859abad473d536.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "e49c942b015d5d4a30c744cae46205e50956262230273fd2a0ee3dce2689f8b2", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b82ffcbdef133ad110332e025391c1a4449eb3a21481ab0a7ea1b3a02af60b75", "generation_seconds_batch": 22.374279395095073, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.022460699081421, "target_z": 0.4843749541967119, "ce": 6.181467056274414, "genuineness": 4.570107460021973, "identity": 0.3045591115951538, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/e49c942b015d5d4a30c744cae46205e50956262230273fd2a0ee3dce2689f8b2.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "e49f55a51b060965e5bd901a0ae71b945e4a0fe73e338e1def56f5f99afce244", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "08256a14f7863d8c869e19cabb171d427683e7c879b2e180a761c735a2a452ab", "generation_seconds_batch": 21.165586818940938, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5954656600952148, "target_z": 0.8593750065433269, "ce": 5.661221981048584, "genuineness": 4.6956281661987305, "identity": 0.6156412959098816, "hypothesis": "He encontrado la cajita. estaba junto a la ventana vieja. podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/e49f55a51b060965e5bd901a0ae71b945e4a0fe73e338e1def56f5f99afce244.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": "e4ccf995278572aa93ea0638c39977d2474a9b2d7be54280df84b6ffb86d0374", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "62721cd0b5ee644019c09c161bab3cfdb076fab799a075e4db334797dca2c5e2", "generation_seconds_batch": 15.662657730979845, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6132571697235107, "target_z": 0.8437499607400387, "ce": 5.764476776123047, "genuineness": 4.7332844734191895, "identity": 0.7579110860824585, "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/vn_STNC_reg_low/e4ccf995278572aa93ea0638c39977d2474a9b2d7be54280df84b6ffb86d0374.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "e52db190fc179ae1921c1585f0af795d891ffbcbaa17ab24bb2fa197a34540e7", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dcd08f65b87b8c0b348dea15d3953f7038faeeca637f3b4bb66029ed1a25bdf4", "generation_seconds_batch": 22.87628270604182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.688870906829834, "target_z": 0.7773436731159092, "ce": 5.879645347595215, "genuineness": 4.507347106933594, "identity": 0.6126354932785034, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/e52db190fc179ae1921c1585f0af795d891ffbcbaa17ab24bb2fa197a34540e7.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "e67c04bef94bb57230fc1aea554424b1408fc0158c3832cc56da0d3b8f0ddee3", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "34908c23c6afbe54d84ef97514ac6290bdfe5a180a85cde610d2ee5c38ec2c3f", "generation_seconds_batch": 16.39742056292016, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5272932052612305, "target_z": 0.04101566671370883, "ce": 5.95112943649292, "genuineness": 4.607763767242432, "identity": 0.7585146427154541, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/e67c04bef94bb57230fc1aea554424b1408fc0158c3832cc56da0d3b8f0ddee3.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "e692b9e6bca06523f8f896a78aeb3464c69b3da8a8679f476c5391d1b4025485", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fd6a616e35162e18680678093e003b22586bf8682c036ef1d8fec753257c6975", "generation_seconds_batch": 22.696449199924245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0380282402038574, "target_z": 0.4707030652921423, "ce": 5.71682071685791, "genuineness": 4.362998008728027, "identity": 0.3742639422416687, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo eh 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/vn_STNC_reg_low/e692b9e6bca06523f8f896a78aeb3464c69b3da8a8679f476c5391d1b4025485.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "e6b017e3a6853facbe4307fb8a215f1b8fee6f52686872a22b7392b33658f469", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ab624867f59d9c43d8668e1f4aeb79cf942cc277a077271cfd6a36984d595d6d", "generation_seconds_batch": 17.54738755291328, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5751078128814697, "target_z": -0.0009765784493592565, "ce": 6.221180438995361, "genuineness": 4.65797233581543, "identity": 0.43937790393829346, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt ?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/e6b017e3a6853facbe4307fb8a215f1b8fee6f52686872a22b7392b33658f469.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "e6b7a47d3c537a869b85f06a3fcaadc816ca187e137abc2951d4a7847e22c599", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "30230bcd44e41d1663cea5dc00af1d550bfe5a1c67ef1ffd5b38ede3941ad797", "generation_seconds_batch": 16.130435936967842, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.147000789642334, "target_z": 0.375000052346615, "ce": 5.927301406860352, "genuineness": 4.250029563903809, "identity": 0.3811386823654175, "hypothesis": "The 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/vn_STNC_reg_low/e6b7a47d3c537a869b85f06a3fcaadc816ca187e137abc2951d4a7847e22c599.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "8a3f31777580edcbaf0c12b03b1a1b88b3496eda8281cc023488c606c47965e0", "generation_seconds_batch": 15.47288943908643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.637377977371216, "target_z": -0.05566413406763769, "ce": 6.38003396987915, "genuineness": 4.419482707977295, "identity": 0.41354790329933167, "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/vn_STNC_reg_low/e6f1e80cc6606ef00fdc322d5829b94fd65ad80a74ab015f88321e5dc194804c.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "e8116bfc69f0a412a14a56d40c1d7c04f82da551b811ff473af6005507ce24f4", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "193c6dc4c9f8e260ff184ae376009e017d0371269166db37acf64ebd9b7eaf3c", "generation_seconds_batch": 23.110999403055757, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2426300048828125, "target_z": 0.29101556202047885, "ce": 6.197352409362793, "genuineness": 4.871357440948486, "identity": 0.8144701719284058, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/e8116bfc69f0a412a14a56d40c1d7c04f82da551b811ff473af6005507ce24f4.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "e81e8d82c9e420895ad3f49bf4889747815f53531194634a666a516d1c142a53", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "4b740e508144e62e56ef28b6d1252842f7a9f3afd4c3691482c8d966da678c09", "generation_seconds_batch": 21.149435841012746, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1684707403182983, "target_z": 1.234374954196712, "ce": 6.046441555023193, "genuineness": 4.369274139404297, "identity": 0.39048779010772705, "hypothesis": "The 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/vn_STNC_reg_low/e81e8d82c9e420895ad3f49bf4889747815f53531194634a666a516d1c142a53.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "e84969809a599fac8fe8a0ab7e56285e4d2fcedda0caeef1bc8da2e229aaff3e", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "441e1e1ffe60ca5bbb84d3c09b4cacefccef8cd3070669b4b96eac4dad0454e5", "generation_seconds_batch": 23.224018677021377, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8712332248687744, "target_z": 0.6171875294449709, "ce": 5.77639102935791, "genuineness": 4.532451152801514, "identity": 0.5082874894142151, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/e84969809a599fac8fe8a0ab7e56285e4d2fcedda0caeef1bc8da2e229aaff3e.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "e8733e23747e5bb109c1db1433e7917fca5e145982ea11fb5a1aff6c0aab9da5", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2fddeb113b5c422f86d875c7b9ce7dd0f06b42278e6eeb9c98f405e335de08d3", "generation_seconds_batch": 22.56491175701376, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9335033893585205, "target_z": 0.5624999738266925, "ce": 5.990842819213867, "genuineness": 4.632867813110352, "identity": 0.775869607925415, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/e8733e23747e5bb109c1db1433e7917fca5e145982ea11fb5a1aff6c0aab9da5.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "e8bca0d164da253971294912b5765603e25cedf2aea88808e27f93f4abbf18a3", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4e21ed837211953b4e1d1d1d9495b1de933cab3fa8e76aaf65a1d136fce961b7", "generation_seconds_batch": 26.791728196083568, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5067219734191895, "target_z": 0.059081949253935104, "ce": 6.0146708488464355, "genuineness": 4.469690799713135, "identity": 0.68075031042099, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/e8bca0d164da253971294912b5765603e25cedf2aea88808e27f93f4abbf18a3.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "e8d3ffbf13e2e74943d33920644ca3a1f9f7bda3785110b68bce8ced73932483", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ccfc3791df7283591db0a1cd638efcb5dffc96a66424933fc72824b68cb2a131", "generation_seconds_batch": 22.56462784006726, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.284115195274353, "target_z": 1.132812470555029, "ce": 5.514282703399658, "genuineness": 4.570107460021973, "identity": 0.5584046244621277, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/e8d3ffbf13e2e74943d33920644ca3a1f9f7bda3785110b68bce8ced73932483.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "e8e96e2563ad7d437b0e6e20d74069bc626569598de0545046372d19733b32dd", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2f23e6f128c5a5c59bf46bc09d5bd9da983b7a4deb0ea14f5145402228ca6d9d", "generation_seconds_batch": 23.148726623039693, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1736879348754883, "target_z": 0.35156258833491283, "ce": 6.006728172302246, "genuineness": 4.507347106933594, "identity": 0.6737576127052307, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/e8e96e2563ad7d437b0e6e20d74069bc626569598de0545046372d19733b32dd.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "e8f73fdf0694aa010113857584740c8836cd15d75347e4bcf9f1337e39201ed1", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "db7e7fb6e0e0d49679f33a51d26882c3ec6655cd51d6077612d430a3303de02a", "generation_seconds_batch": 24.461503827944398, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3641767501831055, "target_z": 1.0624999738266925, "ce": 5.34748649597168, "genuineness": 4.595211505889893, "identity": 0.6762522459030151, "hypothesis": "J'ai trouvé que la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/e8f73fdf0694aa010113857584740c8836cd15d75347e4bcf9f1337e39201ed1.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "e91c42dfdbd67ba97359de71d3ee6f17c9fa67829c849a88c91b69accfffddc4", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "39aa7edbf491a2752efc150c52dce0c3661682c84ba5c95e5c171b9b55636c70", "generation_seconds_batch": 23.182376329088584, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9905561208724976, "target_z": 1.3906249934566732, "ce": 5.418970584869385, "genuineness": 4.7081804275512695, "identity": 0.5965436100959778, "hypothesis": "He encontrado la cajita, estaba junto a la ventana vieja, podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/e91c42dfdbd67ba97359de71d3ee6f17c9fa67829c849a88c91b69accfffddc4.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "e95c72534b64da513edcef0e555d30cb891aa0be27db6eb032934b428c1c6b62", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "ed9252b6cf430610073bc19d2c1246449a92a8a4802195c744f334f785e9fb41", "generation_seconds_batch": 24.322730217012577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8756811618804932, "target_z": 0.6132812156475339, "ce": 5.7128496170043945, "genuineness": 4.406930446624756, "identity": 0.6211128830909729, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/e95c72534b64da513edcef0e555d30cb891aa0be27db6eb032934b428c1c6b62.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "e965cf116a5c145df792c9458c13799435ff09b0dd21d7fb5bd2479a48507084", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a9c4f55eabb6e42cb3ca9c9f655f02f7c77b89062cf51f60d86f137f54d11325", "generation_seconds_batch": 21.17204432899598, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5109562873840332, "target_z": 0.9335937123758705, "ce": 6.0305562019348145, "genuineness": 4.921566009521484, "identity": 0.7474488615989685, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/e965cf116a5c145df792c9458c13799435ff09b0dd21d7fb5bd2479a48507084.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": "e974b7d3fab1e96f1f2801ee83a6186a0059e89c34205ce1f9c3c8c26d905524", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Your message arrived after I had left.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b2bcdb9ca367dc079b759fc40b439c879acffdcdd2d394b292782ae1b0e8fadf", "generation_seconds_batch": 17.504247745964676, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5999135971069336, "target_z": 0.8554686927458899, "ce": 5.808161735534668, "genuineness": 4.369274139404297, "identity": 0.4914730191230774, "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/vn_STNC_reg_low/e974b7d3fab1e96f1f2801ee83a6186a0059e89c34205ce1f9c3c8c26d905524.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": "e9c2f35eb19a9d1a486db695abfda490d92a73eaef22a368378f21412e5025d6", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Someone moved the little blue chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1efa38fa425ce896f9fa7d5b4dd28ec255ea249dbcd25d568427144f5d5f6a70", "generation_seconds_batch": 21.141588732018135, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8578896522521973, "target_z": 0.6289062614508221, "ce": 6.125868320465088, "genuineness": 4.582659721374512, "identity": 0.4426753520965576, "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/vn_STNC_reg_low/e9c2f35eb19a9d1a486db695abfda490d92a73eaef22a368378f21412e5025d6.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "d0ea9b49ef592c474ddd35ff2f9a077b0e33f859dcf30e0530103700e8e7ad11", "generation_seconds_batch": 15.681992769008502, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.063261032104492, "target_z": -0.4296876079648934, "ce": 6.475345611572266, "genuineness": 4.68307638168335, "identity": 0.4890871047973633, "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/vn_STNC_reg_low/e9d266c3907611a53d0c5f7688a7e231bba29ead6f87b8ae8d29cb37ab2fae41.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "eaa3f2e08f865e79943a5d4aa4de0ac3bcecb388e55b070d8e8dbe07c8d47030", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5cc3c42b97a5f69dca6297a3d45aff464a91db48569a1d168556966bacf4cae8", "generation_seconds_batch": 23.895294410991482, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3160197734832764, "target_z": 0.2265624312950678, "ce": 6.062326908111572, "genuineness": 4.344170093536377, "identity": 0.7489877939224243, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/eaa3f2e08f865e79943a5d4aa4de0ac3bcecb388e55b070d8e8dbe07c8d47030.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "eae240ff8e692a714af05ce03c6f5e20c0a745eeb4f167c7da3157c5b43847d0", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c36d1c0b568fc09c47c156cfeced026c076ad9fca9848ca29d86722a70181548", "generation_seconds_batch": 17.952876089024357, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.018012762069702, "target_z": 0.4882812679941489, "ce": 6.125868320465088, "genuineness": 4.909013748168945, "identity": 0.721962571144104, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/eae240ff8e692a714af05ce03c6f5e20c0a745eeb4f167c7da3157c5b43847d0.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": "eaf4a5068b17ba25fe33c13b08c1d75d657279364261ef72b5b0fd6dde9d228a", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I wrote the number on the envelope.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1f4a2f3ff027f2d21d229ba77cea6d145e2e147ae7c85d19917b1a2c012804fd", "generation_seconds_batch": 17.52171370398719, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0624914169311523, "target_z": 0.4492187581791586, "ce": 5.744620323181152, "genuineness": 4.250029563903809, "identity": 0.4249421954154968, "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/vn_STNC_reg_low/eaf4a5068b17ba25fe33c13b08c1d75d657279364261ef72b5b0fd6dde9d228a.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "eb25121c6b23f283ee8e5e8e090e2a803baa0207a8c04ad641a5e5d6ae75864e", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e50146a9ae676b41ed349bee760c1513b9a55ebd3f4d923a83946165478d9ed7", "generation_seconds_batch": 16.496118741924874, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5991441011428833, "target_z": 1.734374954196712, "ce": 5.959072113037109, "genuineness": 4.557555198669434, "identity": 0.7152426242828369, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/eb25121c6b23f283ee8e5e8e090e2a803baa0207a8c04ad641a5e5d6ae75864e.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "ebdd3ba0baa51a339ff5c60a885d7cbb4bca264267913892feeb156a5131599e", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c99b674e1c8642890a208ba2b632298895e5c03cd8b8dff8cfced77e9259c034", "generation_seconds_batch": 23.089595869998448, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.191479444503784, "target_z": 0.33593754253162467, "ce": 6.0146708488464355, "genuineness": 4.494794845581055, "identity": 0.6409733295440674, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ebdd3ba0baa51a339ff5c60a885d7cbb4bca264267913892feeb156a5131599e.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "fd4999123dca2379749d702790b76d42272d5bc29df3cd26d1f3c3aad29bea3f", "generation_seconds_batch": 16.457021304988302, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5384128093719482, "target_z": 0.03125009160657624, "ce": 6.109982967376709, "genuineness": 4.419482707977295, "identity": 0.8877331018447876, "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/vn_STNC_reg_low/ebe89981a89e9506d02bacdfa583f9bfe1f04faa699193749ab6419de85c588f.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "ebe9190f4ab77324ea3d52f3e9233a51c25ce647600492c5e7e1848971912c57", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2550929d6689b6c4822ce8b17e5bcb00cc908aeba6bfc5c6f1860b153d5eda0b", "generation_seconds_batch": 16.47657114895992, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6532878875732422, "target_z": 0.8085937647224855, "ce": 5.879645347595215, "genuineness": 4.632867813110352, "identity": 0.5992282629013062, "hypothesis": "He encontrado la cajita, estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ebe9190f4ab77324ea3d52f3e9233a51c25ce647600492c5e7e1848971912c57.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "ec08651391e55242b1cf1ade8c8ba4512b480fbeedab60b5004865891b1174df", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "208529415149c75a69be7c7a9a0f79f527cf6e79c0afcb9e9d55bd4794667a83", "generation_seconds_batch": 16.76424115197733, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.035804271697998, "target_z": 0.4726562221908608, "ce": 6.109982967376709, "genuineness": 4.758388519287109, "identity": 0.7983444929122925, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ec08651391e55242b1cf1ade8c8ba4512b480fbeedab60b5004865891b1174df.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "ec0bd331e12b18c198a40847a799d23d6ab9e3f7e80b3267b29498a4b2ea358a", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "992da3189a028172bdbfad18082ebb9adfb56484d825eb2d64a20eb4cc4ec59a", "generation_seconds_batch": 16.9289269100409, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.639944314956665, "target_z": 0.8203124967283366, "ce": 5.808161735534668, "genuineness": 4.68307638168335, "identity": 0.6818056106567383, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ec0bd331e12b18c198a40847a799d23d6ab9e3f7e80b3267b29498a4b2ea358a.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "ec242786af44c7bc366972e42b3be2c436d367cfc4a2c8f14036dd6f1b9b5440", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5f3678c8e64aa9d811578ee50abb7bb0ad39957d741b0c074040205674b254e9", "generation_seconds_batch": 16.766530635999516, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0313563346862793, "target_z": 0.4765625359882978, "ce": 6.054384231567383, "genuineness": 4.68307638168335, "identity": 0.6544222831726074, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ec242786af44c7bc366972e42b3be2c436d367cfc4a2c8f14036dd6f1b9b5440.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "ec3463aad4b544ed89eea23c02256f0f91c9a4c783f13ae02b6f17d6fc64fe06", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b58775cd573fa6efa278a15eea4007f4f8e7de94a3e20aef1ebeddf50a14cf8d", "generation_seconds_batch": 16.704489166033454, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7377972602844238, "target_z": 0.7343750588899419, "ce": 6.149696350097656, "genuineness": 4.720732688903809, "identity": 0.5875387787818909, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ec3463aad4b544ed89eea23c02256f0f91c9a4c783f13ae02b6f17d6fc64fe06.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "eccc372f3bba7264c16cbe5b01b4e67de83a50167fd24c860da7dd89904080ca", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ddd0bcf159240d438da3243440324a7be404de6b14eaed1a19d6f1ba6df176bd", "generation_seconds_batch": 22.34314983896911, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0002212524414062, "target_z": 0.503906313797437, "ce": 6.086154937744141, "genuineness": 4.369274139404297, "identity": 0.6787235736846924, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschalten.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/eccc372f3bba7264c16cbe5b01b4e67de83a50167fd24c860da7dd89904080ca.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "ed0c056d4252eba0ae61ff01477e72ca22c82fb05bcbdc9fdf32b4a6e95597a5", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "868a67e34137c023e80153ca9a950ae97014d2568d08a8d1decfd42a7180279e", "generation_seconds_batch": 15.342324733966962, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.639944314956665, "target_z": 0.8203124967283366, "ce": 5.967014789581299, "genuineness": 4.7081804275512695, "identity": 0.5402699708938599, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ed0c056d4252eba0ae61ff01477e72ca22c82fb05bcbdc9fdf32b4a6e95597a5.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "ed417e35140b29f7af7737b6889adda3b7c5116bec0eabd26d8596e837bb7838", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "670fa988928fc8616658b380b010f7aad4a34cd84612e412d7a7f4d5474258a9", "generation_seconds_batch": 17.497616717009805, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.646829605102539, "target_z": -0.06396484150073138, "ce": 6.316492557525635, "genuineness": 4.507347106933594, "identity": 0.5757566690444946, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ed417e35140b29f7af7737b6889adda3b7c5116bec0eabd26d8596e837bb7838.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "eda1f6830b031f3dd7c7f35016969bd275a5efbf403d1a6f184a968a750135b7", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "47e826bd413266ad26e8506ad9b54f05b43cb3e15e0e163561f0b43a80fe4ae7", "generation_seconds_batch": 16.826629890943877, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.563561201095581, "target_z": 1.7656249411100582, "ce": 5.708878040313721, "genuineness": 4.833701133728027, "identity": 0.5987335443496704, "hypothesis": "La porte est encore ouverte j'ai posé ton manteau sur la chaise. Dis-moi quand tout est prêt.", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/eda1f6830b031f3dd7c7f35016969bd275a5efbf403d1a6f184a968a750135b7.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "edf986dcc9bc7a2075403aadf38d7fc676aa644b9fb3c98ae165059d359a4590", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "951615c69fa6e85794b9a7b16dcce1bc3a59d7f3f18e5de92cc89bf70489f77e", "generation_seconds_batch": 16.86311360099353, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1558966636657715, "target_z": 0.36718742475174093, "ce": 6.252951145172119, "genuineness": 4.7834930419921875, "identity": 0.5234167575836182, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/edf986dcc9bc7a2075403aadf38d7fc676aa644b9fb3c98ae165059d359a4590.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "ee4a54d004147dbbec1aa074f58da621f5c5e8341987700f0949ddf2a375d597", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "351a056a0250a57af280d5345c3d40e9dbc9778d2d06fafe13dcce245b2b3ca1", "generation_seconds_batch": 17.098108747974038, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7644845247268677, "target_z": 0.7109374901850097, "ce": 6.133810997009277, "genuineness": 4.808597087860107, "identity": 0.5756643414497375, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ee4a54d004147dbbec1aa074f58da621f5c5e8341987700f0949ddf2a375d597.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "ef7301197b32c1e56e81fc5e0d0d53ab489b0fb3d770acdb086dae7f6fc783d4", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4f8d63e9b3ecd5246b27382f4ecf98eebf28e75b30d2f024a9121b221d44b806", "generation_seconds_batch": 17.73909819102846, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5056099891662598, "target_z": 0.06005852770329436, "ce": 6.165581703186035, "genuineness": 4.2374773025512695, "identity": 0.3870362937450409, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ef7301197b32c1e56e81fc5e0d0d53ab489b0fb3d770acdb086dae7f6fc783d4.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "ef93133a54e6e6d9a50ce44f4e86c62a8bf796f9388c24c1ed9ab75b77e98189", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "64782b22d24a19e87d63f19e55e408a11c4fa3e6b4976b102a80a0e175a38c1d", "generation_seconds_batch": 16.605105969007127, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.444238305091858, "target_z": 0.992187477098356, "ce": 5.911416053771973, "genuineness": 4.808597087860107, "identity": 0.7731480002403259, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ef93133a54e6e6d9a50ce44f4e86c62a8bf796f9388c24c1ed9ab75b77e98189.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "f007e904a7fc85e4032684111a544b318d5b4a4a11b4b66c5cf9bb2e836579c6", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "811790802b1a5960b501ff64d05a6be42555d381200ba32126c03bf8d8627814", "generation_seconds_batch": 17.43679237004835, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9690862894058228, "target_z": 0.5312499869133462, "ce": 6.205295085906982, "genuineness": 4.325342178344727, "identity": 0.3757677376270294, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f007e904a7fc85e4032684111a544b318d5b4a4a11b4b66c5cf9bb2e836579c6.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "f09210330857618b3e701ddeae70422903944f85038269f001feca93c36e9c26", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2c19c18f7317e0a37abdc78ad01dcc4f2180d09746cf34f936ca4c4751bd0ac6", "generation_seconds_batch": 20.91106655402109, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.221845030784607, "target_z": 1.1875000261733075, "ce": 6.006728172302246, "genuineness": 4.532451152801514, "identity": 0.3046770691871643, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f09210330857618b3e701ddeae70422903944f85038269f001feca93c36e9c26.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "f0963e13642af2ed6c8bd7b3817eca6d1a46575c7ba74208fb28016b0bf08350", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "623a33c1e28b7e9b67ba04db5ab9f2d1155b1da8f414b338f9a0b0615dd90f25", "generation_seconds_batch": 23.43883040198125, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.862337589263916, "target_z": 0.624999947653385, "ce": 5.700935363769531, "genuineness": 4.444586753845215, "identity": 0.6842848062515259, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f0963e13642af2ed6c8bd7b3817eca6d1a46575c7ba74208fb28016b0bf08350.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "f09db9b3e1b2c90343b99e0a02ec7998177d7c7029b39201f141d25bbdbb6f3b", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5d1d1fd38bb0c8f47e75223ef4ce60afba791bd9d6ca599f48695fc51e315ff0", "generation_seconds_batch": 22.784538623993285, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.466135025024414, "target_z": 0.094726643882628, "ce": 5.998785495758057, "genuineness": 4.369274139404297, "identity": 0.6786035299301147, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f09db9b3e1b2c90343b99e0a02ec7998177d7c7029b39201f141d25bbdbb6f3b.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "f12bf57575d03a2366f84c5449efa7b6ec67e5e0c44da34cd26a2d612cb8c12a", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f3aadc7e8690d46d818b66e59767af42df9b9ef8e1b0e2f26b68cae4abb0872e", "generation_seconds_batch": 26.506137130898423, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0091171264648438, "target_z": 0.496093686202563, "ce": 6.284721851348877, "genuineness": 4.545003414154053, "identity": 0.6465923190116882, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f12bf57575d03a2366f84c5449efa7b6ec67e5e0c44da34cd26a2d612cb8c12a.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "f151f3ae81f3b3ab8b25815ae56dcaee8109dab2d6eb5ccf54266ef510ebd448", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ba24504d08a9b67be3d7d691e48f0921fea833992ded9d760d7e69b85c345a11", "generation_seconds_batch": 17.178188227000646, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.160344362258911, "target_z": 0.3632813203407639, "ce": 6.133810997009277, "genuineness": 4.306513786315918, "identity": 0.2438952773809433, "hypothesis": "The 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/vn_STNC_reg_low/f151f3ae81f3b3ab8b25815ae56dcaee8109dab2d6eb5ccf54266ef510ebd448.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "f176856a80c27ec293f65678337487e0591d38414bd1a57557111ac5860de170", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a3112c3b913c0f00455222aa57dea46163386540b46f8cede244d36901cc5f63", "generation_seconds_batch": 17.099823054042645, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5724568367004395, "target_z": 1.7578125229016441, "ce": 5.637393951416016, "genuineness": 4.457139015197754, "identity": 0.2854568362236023, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f176856a80c27ec293f65678337487e0591d38414bd1a57557111ac5860de170.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "f1aa40df249b9a9fcf29bcc1cda3620806edb3db3b856211791c32854059ba02", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eda8856fcd55c4c2fd07b9b35263efe0b4ee718f5e20fafa1b7462bf6bf2464d", "generation_seconds_batch": 24.24937932402827, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.426446795463562, "target_z": 1.0078125229016441, "ce": 5.967014789581299, "genuineness": 4.720732688903809, "identity": 0.6871734857559204, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f1aa40df249b9a9fcf29bcc1cda3620806edb3db3b856211791c32854059ba02.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "f1ce94b972905e53153dbb361f978c4437489c23f428d656053f56ab97a9824e", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fb4dec9cb8d78b3a699d79e0d06c8e44736691b1d13525d46f1f378d3c0609e5", "generation_seconds_batch": 22.12175811594352, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1773663759231567, "target_z": 1.226562535988298, "ce": 6.221180438995361, "genuineness": 4.758388519287109, "identity": 0.6450822353363037, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f1ce94b972905e53153dbb361f978c4437489c23f428d656053f56ab97a9824e.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "f205c1cccbfe33bf0796c5a27ba1934e4944b134149578bde274c7e968e3d576", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "efceda637bdfde3ddb59afb718d84c1640b1676efef0dc65effa1be883ed11d5", "generation_seconds_batch": 16.617920999997295, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9112640619277954, "target_z": 0.5820312287341877, "ce": 5.990842819213867, "genuineness": 4.2249250411987305, "identity": 0.49665454030036926, "hypothesis": "De deur staat nog open. De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.2631578947368421, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f205c1cccbfe33bf0796c5a27ba1934e4944b134149578bde274c7e968e3d576.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "f207f0e53f34da2a75ef4373beb25f4387856a8888d4e141cb395128fd6951be", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ccd54bae450139ab2654e8c3ca4997986d6fb9b9d5cb796af72951e3ba63bb58", "generation_seconds_batch": 21.904145609005354, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5106136798858643, "target_z": 0.05566413406763769, "ce": 6.213237762451172, "genuineness": 4.582659721374512, "identity": 0.6246978640556335, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f207f0e53f34da2a75ef4373beb25f4387856a8888d4e141cb395128fd6951be.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "4bde7d3881d27ebf1d0f176304f2a705ac3c976ea4674bc723f2b3049bcf1443", "generation_seconds_batch": 15.356928450986743, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.018012762069702, "target_z": 0.4882812679941489, "ce": 5.935244083404541, "genuineness": 4.582659721374512, "identity": 0.6775057911872864, "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/vn_STNC_reg_low/f21bc0adf2df6aeda4aa5bc68657c60b57559ffed3b647118e970ebfd0ba6ad8.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "f2383ed6d062cc999bbab25448b498471de94a5f7f518a203e9740568db010b0", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8843dceed2046e2a01d791dcf9ebc76fcb3b2b67c74465c77c8e14c1906e1b5d", "generation_seconds_batch": 17.937688520993106, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0869545936584473, "target_z": 0.4277344510661749, "ce": 6.181467056274414, "genuineness": 4.7834930419921875, "identity": 0.7210274934768677, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f2383ed6d062cc999bbab25448b498471de94a5f7f518a203e9740568db010b0.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "f24b3bee8f0db43b4652b280df9525cfe2dd79a1a78e8127e216852f927b61c7", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b5e8d4246acce51c5f156512485cb2d259b402938bb1f3217e52281bcb95ed84", "generation_seconds_batch": 23.289206663961522, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5509870052337646, "target_z": 0.8984375163583171, "ce": 5.546053409576416, "genuineness": 4.846253395080566, "identity": 0.7475287914276123, "hypothesis": "He encontrado la cajita, estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f24b3bee8f0db43b4652b280df9525cfe2dd79a1a78e8127e216852f927b61c7.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "f266ee6ae238c9d590eca462c3436463f6f928a58f057a52653d7ef61b5277cc", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "69935178af659faeb9ba2f369b33c5f903f91b77c85d4f977c73dad2fba0c675", "generation_seconds_batch": 16.374500519013964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4928224086761475, "target_z": 0.07128897048446582, "ce": 5.998785495758057, "genuineness": 4.494794845581055, "identity": 0.46160638332366943, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f266ee6ae238c9d590eca462c3436463f6f928a58f057a52653d7ef61b5277cc.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "f27bbfd72a67105fcdd4adfaa465ed587c497e0634ec34409f3c99977f9e815b", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b0201d57413effce3e4bc1fd54ab4894a6136fc4068ca75ad57f94e39f45c875", "generation_seconds_batch": 21.678141291020438, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.488716959953308, "target_z": 0.9531249672833656, "ce": 5.363371849060059, "genuineness": 4.362998008728027, "identity": 0.7135559320449829, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre.", "wer": 0.3684210526315789, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f27bbfd72a67105fcdd4adfaa465ed587c497e0634ec34409f3c99977f9e815b.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "f2899a78a1d3d33b148d443e51abfb591252cea443a7b71022bf946902ff8e32", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7f1db88ce51b2cf6622072dbe6248dd4983d42789a7ccbaa7059014de1daff9e", "generation_seconds_batch": 21.199929359019734, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.657393217086792, "target_z": -0.07324212738318434, "ce": 6.109982967376709, "genuineness": 4.582659721374512, "identity": 0.4179689884185791, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f2899a78a1d3d33b148d443e51abfb591252cea443a7b71022bf946902ff8e32.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "f2de1fb8d939e7d897662ca2a33a5ab3daddb464013313208d085739386ae8c6", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d8303589c445bed45115017bd6648dc9c7777d78450f03bfd466ec74fb10a7ec", "generation_seconds_batch": 23.797784519963898, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1862621307373047, "target_z": 1.2187500130866538, "ce": 5.788304805755615, "genuineness": 4.7081804275512695, "identity": 0.757821261882782, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f2de1fb8d939e7d897662ca2a33a5ab3daddb464013313208d085739386ae8c6.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "f2e28bb69cf9e14159b70990e4597d6966c8607f3ebb900532a4b1a69c898421", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2a1380c8c88eb6a9b6e508558987479e5ac37980030048baaf5514575ea1a077", "generation_seconds_batch": 16.408854873036034, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0380282402038574, "target_z": 0.4707030652921423, "ce": 6.022613525390625, "genuineness": 4.7081804275512695, "identity": 0.8032336235046387, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f2e28bb69cf9e14159b70990e4597d6966c8607f3ebb900532a4b1a69c898421.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "f3265f23e10e55793e639f75f12f026213c6819dd92ae48f758b4c888054647b", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "66d086e6125c37b8229482610881694b4b4b85ee719db8032f26732db54a08cd", "generation_seconds_batch": 21.809175229049288, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8979203701019287, "target_z": 0.5937500654332687, "ce": 6.109982967376709, "genuineness": 4.469690799713135, "identity": 0.7028959393501282, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f3265f23e10e55793e639f75f12f026213c6819dd92ae48f758b4c888054647b.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": "f32f0ee0c9669bba7d1e94eddf770a8e8a253d2fda7811b11691bf6bceabe5dd", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Your message arrived after I had left.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "814cf28e4bf1f337ea138d98261faa5518833eb36cfe4573f5a26a0ad5276dfa", "generation_seconds_batch": 22.753210394992493, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.417551040649414, "target_z": 1.015625045803288, "ce": 5.919358730316162, "genuineness": 4.532451152801514, "identity": 0.5518428087234497, "hypothesis": "Your message arrived after I had left. I I read it twice before putting the phone away. Let me know when we can talk.", "wer": 0.043478260869565216, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f32f0ee0c9669bba7d1e94eddf770a8e8a253d2fda7811b11691bf6bceabe5dd.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "f34aed2edf6a508d2c02caebe9c423b41b593920803a39ed3f4bbe880e696ff3", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f657cc5d93830c6a7cad00163c6fdf3da1e3fedecd3e6e78eed98a4c479800af", "generation_seconds_batch": 22.80731120100245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1180896759033203, "target_z": 0.4003906732570357, "ce": 6.157639026641846, "genuineness": 4.3881025314331055, "identity": 0.7347774505615234, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f34aed2edf6a508d2c02caebe9c423b41b593920803a39ed3f4bbe880e696ff3.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "f37f78653458e6018f8b8e78282348d9b6deffa893b3a7467669d70b3170123b", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e84959efd43f1790158fd8af051fc617e76ba9cfaca935d0a0cd8bdca1f7e82a", "generation_seconds_batch": 26.79255271994043, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4397903680801392, "target_z": 0.996093790895793, "ce": 5.442798614501953, "genuineness": 4.557555198669434, "identity": 0.5817433595657349, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f37f78653458e6018f8b8e78282348d9b6deffa893b3a7467669d70b3170123b.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "f3900b16e02e56f040cebd2f4042a09a50e17da15d4c9bd2f6480c46c4187c9b", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9f17b8807da2c59e1cc1cd4abb1d7dec078fbbe9ba184e1727c62eb325d4ece1", "generation_seconds_batch": 22.14416596200317, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.508389949798584, "target_z": 0.05761708157989622, "ce": 6.284721851348877, "genuineness": 4.457139015197754, "identity": 0.666850209236145, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f3900b16e02e56f040cebd2f4042a09a50e17da15d4c9bd2f6480c46c4187c9b.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "f3a3bec828d95e4db09d49cb4bfdccb57dfdca5582e062dfaaf0907f782492c5", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "86f7dbf517e0bafed0779dd2b4645a5e050ac5d1dedbfa9e596801b06a7b5f89", "generation_seconds_batch": 22.43666469398886, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7155580520629883, "target_z": 0.753906209104207, "ce": 5.700935363769531, "genuineness": 4.482243061065674, "identity": 0.6930018663406372, "hypothesis": "La puerta sigue abierta. he dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f3a3bec828d95e4db09d49cb4bfdccb57dfdca5582e062dfaaf0907f782492c5.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "f3b162680f687cf91fb765e2389d377b2ce7796bcf7b3116fc2385e9228ea69f", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f4dbbb11fc699c17d102f0c61648c9dc7249623843d8ad238338b75f59ac8afd", "generation_seconds_batch": 22.854234994971193, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3019065856933594, "target_z": 1.117187529444971, "ce": 5.359400749206543, "genuineness": 4.243753433227539, "identity": 0.5613547563552856, "hypothesis": "The 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/vn_STNC_reg_low/f3b162680f687cf91fb765e2389d377b2ce7796bcf7b3116fc2385e9228ea69f.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "f3b2e41cc826f2f38d5a066fcebf97437b3ff7f119f957cbc206e0318b7f11d7", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "51b36b9ea6b46e5201e145771cd452c35ce89a7f5949a82f545f8d129dba103a", "generation_seconds_batch": 22.50628304295242, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.269317150115967, "target_z": 0.26757809800877663, "ce": 6.046441555023193, "genuineness": 4.331617832183838, "identity": 0.5634029507637024, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f3b2e41cc826f2f38d5a066fcebf97437b3ff7f119f957cbc206e0318b7f11d7.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "f44908d065ae5756456457aa4fdc40def585e5662df6af57667a93ca3db50bdd", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1a7c5b1c06129de1cadfb295887407ee508e80bb235ecf004aef6de38aca08b6", "generation_seconds_batch": 17.099203041987494, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8801289796829224, "target_z": 0.6093750065433269, "ce": 5.990842819213867, "genuineness": 4.632867813110352, "identity": 0.5624085664749146, "hypothesis": "He encontrado la cajita. estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f44908d065ae5756456457aa4fdc40def585e5662df6af57667a93ca3db50bdd.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "f47accc1ee683c7b23861cb70908be8801d7579e7006c9d39c8b225ce1437c5b", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2621ca15b6b2276ff36756129a9cf2dfcbcbf7a9cbf4caf409f160bf76a64bcd", "generation_seconds_batch": 23.789345577941276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6799750328063965, "target_z": 0.7851563007107832, "ce": 5.998785495758057, "genuineness": 4.545003414154053, "identity": 0.5885017514228821, "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/vn_STNC_reg_low/f47accc1ee683c7b23861cb70908be8801d7579e7006c9d39c8b225ce1437c5b.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "f5070eb273d6de9cdbee1d45d456a09950b826775940abf9d57be78d379b2250", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f07f143d7264300a445eecd9c9f0c728c5421e8cd46b17f26b24916807f525b5", "generation_seconds_batch": 21.74014264100697, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6221528053283691, "target_z": 0.8359375425316247, "ce": 5.935244083404541, "genuineness": 4.883909225463867, "identity": 0.7065954208374023, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f5070eb273d6de9cdbee1d45d456a09950b826775940abf9d57be78d379b2250.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "f50ad83d1bac7bfd3bdcd0151210d0d1efbe7d3c8face319e2b3237f5e9b381a", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8d7061cb2306b72f331366c549b6e34902ca534af51c4abdc60c96d1ffddae01", "generation_seconds_batch": 22.418045150930993, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8178589344024658, "target_z": 0.6640624574683753, "ce": 6.1020402908325195, "genuineness": 4.6705241203308105, "identity": 0.42123010754585266, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f50ad83d1bac7bfd3bdcd0151210d0d1efbe7d3c8face319e2b3237f5e9b381a.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "f51a380fcfc64fff23715db9b77165d0e19debd64f6764882cc5a74170ca4be5", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7673a9023d019282028bd37f2569e005a7314c31a74a28870a21feaf81cb97cc", "generation_seconds_batch": 16.957680628984235, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.102522134780884, "target_z": 0.4140625621616053, "ce": 6.292664527893066, "genuineness": 4.720732688903809, "identity": 0.6593714952468872, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f51a380fcfc64fff23715db9b77165d0e19debd64f6764882cc5a74170ca4be5.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "f59a30fc59c07a171f8630e23cea444c71d612b352df2626dfa09f9626d58e06", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "62b4816434fab47d219fd94d622c2e41a3ca827c106e03d688ae49d236166475", "generation_seconds_batch": 16.535791578004137, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9468469619750977, "target_z": 0.5507812418208414, "ce": 6.0305562019348145, "genuineness": 4.632867813110352, "identity": 0.7559638023376465, "hypothesis": "He encontrado la cajita, estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f59a30fc59c07a171f8630e23cea444c71d612b352df2626dfa09f9626d58e06.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "f5a08ccb5e4ea08c6fea3638d76c49f376035a5367113d28e8db8cbb82b31922", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "732496e72ef853cdee65d660d60e0acfb7aa7dd378507b95d7d96309d2d23d04", "generation_seconds_batch": 23.182376329088584, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.221845030784607, "target_z": 1.1875000261733075, "ce": 5.681078910827637, "genuineness": 4.821148872375488, "identity": 0.7125666737556458, "hypothesis": "He encontrado la cajita, estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f5a08ccb5e4ea08c6fea3638d76c49f376035a5367113d28e8db8cbb82b31922.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "f6de5194b1a5e3bc4cc00385351fea1569e01490a8b6f815f4e469c24ee5e27f", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "920309dcffde29cbe24a5ced69dcb8a69444926f5739b132693dfe57bf7d47e7", "generation_seconds_batch": 22.02541480807122, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.848224401473999, "target_z": 1.515625045803288, "ce": 5.752562999725342, "genuineness": 4.6705241203308105, "identity": 0.6540119647979736, "hypothesis": "Ich hab die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f6de5194b1a5e3bc4cc00385351fea1569e01490a8b6f815f4e469c24ee5e27f.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "f71929716a99218a15ca246853582e94dc08bc704fa9cc7ad86dade9934796ef", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "543ed5a79ce9472a552bef10f8102df9fe131b9ade0bae132df19a2a85c1aceb", "generation_seconds_batch": 22.61034290806856, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.428328275680542, "target_z": 0.12792968300146276, "ce": 5.736677646636963, "genuineness": 4.331617832183838, "identity": 0.584942102432251, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f71929716a99218a15ca246853582e94dc08bc704fa9cc7ad86dade9934796ef.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "f7646f781b9950c8ca35acdb23ea8ca9cf6c15f32ee014e7f50fdbf6d1984025", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e95e543a58d8a6b72f768820722d110f252827291083a985c77a77bba19f7c4d", "generation_seconds_batch": 17.23874403501395, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6629531383514404, "target_z": -0.07812501962998063, "ce": 6.046441555023193, "genuineness": 4.356721878051758, "identity": 0.7577276229858398, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f7646f781b9950c8ca35acdb23ea8ca9cf6c15f32ee014e7f50fdbf6d1984025.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "f7836f9bdc8815a7d36c4dc095b2c5242657ce315bafae7b03237076b0f21e3a", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "35c436b9aca014231f4689ac658a7f3fb51f6bbc82741016ca72b8f8a5855c21", "generation_seconds_batch": 17.099823054042645, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2574279308319092, "target_z": 1.1562500392599613, "ce": 5.80021858215332, "genuineness": 4.344170093536377, "identity": 0.614291787147522, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f7836f9bdc8815a7d36c4dc095b2c5242657ce315bafae7b03237076b0f21e3a.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "b2cb92a32e003973ba9f80af0c43d8e7089c4977f07e5d99aa3ccf661b16e59f", "generation_seconds_batch": 16.74950994201936, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.025454044342041, "target_z": -0.3964843594595987, "ce": 6.117925643920898, "genuineness": 4.65797233581543, "identity": 0.6870548725128174, "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/vn_STNC_reg_low/f7a65885d0e107603bdce9d92b5fc3b68a68524db7b0b06f79b6538794e8bf5c.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "f7ec89dc0042e5c1eb252bb74df35b5ac5da42617ead2bca987a15b13dcea472", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "800bd5dc1639354d1bf44bdc7dd5b5f2922c2c44c436954cd52d347e02d2273a", "generation_seconds_batch": 22.236689200042747, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8801289796829224, "target_z": 0.6093750065433269, "ce": 6.125868320465088, "genuineness": 4.645420074462891, "identity": 0.5752512216567993, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f7ec89dc0042e5c1eb252bb74df35b5ac5da42617ead2bca987a15b13dcea472.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": "f848377a8deb0be6024a19595333cfcb7cba11bc2eff79d93ce6d656234c8924", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1c60dbcbf5d27de79f5836bc7a1f9aa3b3c821efd50fbb4ba9ce8eb538fbcee4", "generation_seconds_batch": 15.845409853965975, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1239920854568481, "target_z": 1.273437464011702, "ce": 5.319687366485596, "genuineness": 4.720732688903809, "identity": 0.8377701640129089, "hypothesis": "J'ai écrit le numéro sur l'enveloppe, les biens et verifi les deux derniers chiffres. Appelle moi quand you're a moment.", "wer": 0.3181818181818182, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f848377a8deb0be6024a19595333cfcb7cba11bc2eff79d93ce6d656234c8924.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": "vn_STNC_reg_low", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": "86ec62cb5cb9db99e186638dba55906a67cc7fbf620e60114d6ea0c37850cefd", "generation_seconds_batch": 16.59049503400456, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.511725664138794, "target_z": 0.054687555618278434, "ce": 6.260893821716309, "genuineness": 4.469690799713135, "identity": 0.6138999462127686, "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/vn_STNC_reg_low/f86feddc7af334e5c3d6691e715ece3168f8e7618bd3b63ca7d0171f412fdf9d.mp3" }, { "id": "vn_STNC_reg_low-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-1", "key": "f9bd3fce65d3c8b57f4586fa34d3d97120063615c26d71bc470e086b4bb33a7e", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "71da054de4ab8b97c80eb17b95891eda02f605ea0d28213cf490033c84da8635", "generation_seconds_batch": 23.313514395966195, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9193903207778931, "target_z": 1.4531249672833657, "ce": 5.80021858215332, "genuineness": 4.6705241203308105, "identity": 0.6619490385055542, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f9bd3fce65d3c8b57f4586fa34d3d97120063615c26d71bc470e086b4bb33a7e.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "f9f7a2e420394eeebdf5d707ee9120f5a64f6d817a3bc2899eb6f05ec9c52489", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "46010bddfbed9101ed7f8b7c3ca5048d53eec61c8520cd56ba37f3f8cf29c1b3", "generation_seconds_batch": 23.043923992081545, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2804367542266846, "target_z": 0.2578125229016441, "ce": 6.197352409362793, "genuineness": 4.557555198669434, "identity": 0.6037155985832214, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/f9f7a2e420394eeebdf5d707ee9120f5a64f6d817a3bc2899eb6f05ec9c52489.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": "fa03a2741afa172e7d417adfce6b1839c33325a505ec7cb5bb1c5f0536f74dfa", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9fe82c21cc7cac5414f163e1cababcdc0e2403225203055813dce4c9a0613b0b", "generation_seconds_batch": 16.408821127959527, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8890247344970703, "target_z": 0.6015624836416829, "ce": 6.022613525390625, "genuineness": 4.394378662109375, "identity": 0.6991220116615295, "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/vn_STNC_reg_low/fa03a2741afa172e7d417adfce6b1839c33325a505ec7cb5bb1c5f0536f74dfa.mp3" }, { "id": "vn_STNC_reg_low-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-1", "key": "fa3be5c098e3ae9224273e5d5b52b64fbc90ce2dc2ee13b4c55357eaa60a5361", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "bd1419ed31ab5078d906b8ced0e4281e6a213ac7ed42cdea2eda528b698d3e19", "generation_seconds_batch": 23.951120762038045, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.079513430595398, "target_z": 1.3124999738266925, "ce": 5.764476776123047, "genuineness": 4.833701133728027, "identity": 0.5533576011657715, "hypothesis": "He encontrado la cajita, estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que 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/vn_STNC_reg_low/fa3be5c098e3ae9224273e5d5b52b64fbc90ce2dc2ee13b4c55357eaa60a5361.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "fa82ad80ddfcc0bd8942247b60f14f9fbad808c284b2e007aa889805078c4a41", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La puerta sigue abierta.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9e41e1e95fdf9aa370de7cf61a6cb08dca20d2f619fdff16a605517717745e37", "generation_seconds_batch": 16.563936572987586, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.417208671569824, "target_z": 0.13769525810859537, "ce": 5.831989288330078, "genuineness": 4.2751336097717285, "identity": 0.8077428340911865, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/fa82ad80ddfcc0bd8942247b60f14f9fbad808c284b2e007aa889805078c4a41.mp3" }, { "id": "vn_STNC_reg_low-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-es-0", "key": "fa924c29f1f97c416510dcdb5a5e8496e82a0c2ed4528c4763708dc82953979c", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2b001da53ade69c7a4cf90cb74265fc1d0ab4d73ac0391a1dd07459881c386be", "generation_seconds_batch": 20.126902670017444, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.595123291015625, "target_z": -0.01855478115136589, "ce": 5.927301406860352, "genuineness": 4.469690799713135, "identity": 0.6988323926925659, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/fa924c29f1f97c416510dcdb5a5e8496e82a0c2ed4528c4763708dc82953979c.mp3" }, { "id": "vn_STNC_reg_low-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-1", "key": "fb0ca749876cb63c3769b0e608f526b23908be8f9580a73f1d50a64e9d740f20", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8280580877a4b7fc9665ed52bc5a21604ba03e1ee9465ed2d97ecbb813585183", "generation_seconds_batch": 17.214738643961027, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4049770832061768, "target_z": 0.1484374116650872, "ce": 6.022613525390625, "genuineness": 4.444586753845215, "identity": 0.5841289758682251, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/fb0ca749876cb63c3769b0e608f526b23908be8f9580a73f1d50a64e9d740f20.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "fb4626e4fe998f342c063a339751e544f146f3653a113d68795abb1c9ca70d1b", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3f0264e053a0fac7793d7bf0f277fe781d87204ed7ac86120d5ff20de628eeab", "generation_seconds_batch": 22.499867534963414, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8178589344024658, "target_z": 0.6640624574683753, "ce": 6.0305562019348145, "genuineness": 4.375550270080566, "identity": 0.7920472025871277, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/fb4626e4fe998f342c063a339751e544f146f3653a113d68795abb1c9ca70d1b.mp3" }, { "id": "vn_STNC_reg_low-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-0", "key": "fd1481860a1abc0a28ba793226daba039d2ff933183af909893e0e6172928fad", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7308ebb8fae38a84767ad6e63a2980fe4c20c296371a169d6a8e33f369b08f06", "generation_seconds_batch": 16.704489166033454, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6844229698181152, "target_z": 0.7812499869133462, "ce": 6.1020402908325195, "genuineness": 4.6956281661987305, "identity": 0.4670812487602234, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/fd1481860a1abc0a28ba793226daba039d2ff933183af909893e0e6172928fad.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": "fd5acc0a4aa55c4e2bf84740e9a3edc2d603d3829d0a8e6759410ef1890f8884", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I wrote the number on the envelope.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ddb1ad90c56928a853cc5c822d43d8a8411afeae4613b036edb3255f7eba20aa", "generation_seconds_batch": 21.070798308006488, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.608809232711792, "target_z": 0.8476562745374758, "ce": 5.9431867599487305, "genuineness": 4.400654315948486, "identity": 0.6162240505218506, "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/vn_STNC_reg_low/fd5acc0a4aa55c4e2bf84740e9a3edc2d603d3829d0a8e6759410ef1890f8884.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": "fe64091024a6f3465a351062a7fefdb88581437b4c01d132a6ad2537e8d16708", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Someone moved the little blue chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "73c72e1e8d1b72701dc07ad2deb53975b64ae1c72e33a87730c7a9ebc3fb42a4", "generation_seconds_batch": 17.476938255946152, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.791171669960022, "target_z": 0.6875000261733075, "ce": 5.816103935241699, "genuineness": 4.607763767242432, "identity": 0.5211928486824036, "hypothesis": "Someone moved the little blue chair. He used to stand here beside the tall bookshelf. We can put it back after lunch.", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/fe64091024a6f3465a351062a7fefdb88581437b4c01d132a6ad2537e8d16708.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": "feeafed73afb126635b3f2f2c21b8e3fdb4b32b353a699bed15b8da6d35c7a3d", "bucket": "vn_STNC_reg_low", "phase": "heldout", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The keys are on the kitchen table.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fcb1f5c569924ed646abdd133bfd1523fc75649233e463ea7e01dcc32e1751fa", "generation_seconds_batch": 17.197381179081276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.488716959953308, "target_z": 0.9531249672833656, "ce": 5.764476776123047, "genuineness": 4.570107460021973, "identity": 0.3038524389266968, "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/vn_STNC_reg_low/feeafed73afb126635b3f2f2c21b8e3fdb4b32b353a699bed15b8da6d35c7a3d.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "ff5f2b377ce42acd92ef932ec67ef52a2876c24c1d66afc18e0dc7415e2728ce", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f96c9c962a2a008732d9b77ce6c890366180e3da6dd3851a84e57603cab875dc", "generation_seconds_batch": 22.94549871806521, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7733802795410156, "target_z": 0.7031249672833656, "ce": 6.062326908111572, "genuineness": 4.344170093536377, "identity": 0.5246114134788513, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ff5f2b377ce42acd92ef932ec67ef52a2876c24c1d66afc18e0dc7415e2728ce.mp3" }, { "id": "vn_STNC_reg_low-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-1", "key": "ff6fae24833c2b77c478539952fb3304ed6b1f3d1d1d8620768a012c69f3bf37", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "93cc1807f4807ef93a855824cf5daddecc5c4b3754a5d59d9cdd344adf6788be", "generation_seconds_batch": 21.904287863057107, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0936264991760254, "target_z": 0.42187498037001936, "ce": 6.3641486167907715, "genuineness": 4.632867813110352, "identity": 0.4213404357433319, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ff6fae24833c2b77c478539952fb3304ed6b1f3d1d1d8620768a012c69f3bf37.mp3" }, { "id": "vn_STNC_reg_low-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-de-0", "key": "ff70c704522e63451fb39680cdb9a485d74ad3f4b7fb7786dc5f5f855676ebf4", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7c36491b4fbe3becc455274d6ba4584bcffd1674c21658cd907f4279edc174c1", "generation_seconds_batch": 16.20014598907437, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7185513973236084, "target_z": -0.1269531045521035, "ce": 6.109982967376709, "genuineness": 4.344170093536377, "identity": 0.482475608587265, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ff70c704522e63451fb39680cdb9a485d74ad3f4b7fb7786dc5f5f855676ebf4.mp3" }, { "id": "vn_STNC_reg_low-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-fr-1", "key": "ff8b220971d8d179d19fe07528dab9ae4b44e8d906db061f771154743ec18596", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "924444d5d3fa0091ff63642b1124fb64ab45a0df3deab417366c9f74496b10a1", "generation_seconds_batch": 16.767505232943222, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3916335105895996, "target_z": 0.16015614367093828, "ce": 6.1020402908325195, "genuineness": 4.770940780639648, "identity": 0.8105312585830688, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ff8b220971d8d179d19fe07528dab9ae4b44e8d906db061f771154743ec18596.mp3" }, { "id": "vn_STNC_reg_low-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-en-0", "key": "ff8ecda4bcb240c3ab9eba6049cbce2a5068d0907d59e58a4979f8865e92cde7", "bucket": "vn_STNC_reg_low", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] The door is still open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] I put your coat on the chair.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5834bb456a78de6dd4b74e4c074491adfbf181db61deb6d5446ad41f14639493", "generation_seconds_batch": 16.35425183805637, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8534417152404785, "target_z": 0.6328125752482591, "ce": 6.284721851348877, "genuineness": 4.494794845581055, "identity": 0.47522062063217163, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ff8ecda4bcb240c3ab9eba6049cbce2a5068d0907d59e58a4979f8865e92cde7.mp3" }, { "id": "vn_STNC_reg_low-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_STNC_reg_low-nl-0", "key": "ffae9fe4677e3315e7e68bf6102348ac107abb3e7096b470e38b3234ee8b939b", "bucket": "vn_STNC_reg_low", "phase": "grid", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Meek and deferential interpersonal stance, with tentative attacks and a yielding, submissive attitude. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0f2ae76f2db52975f6c0462967fe8e7911722753011d37336ec88ad4125ee736", "generation_seconds_batch": 16.365986089920625, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7022144794464111, "target_z": 0.7656249411100581, "ce": 5.959072113037109, "genuineness": 4.394378662109375, "identity": 0.5933618545532227, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_STNC_reg_low/ffae9fe4677e3315e7e68bf6102348ac107abb3e7096b470e38b3234ee8b939b.mp3" }, { "key": "original-vn_STNC_reg_low-de-0-seed1701-scale0.0", "bucket": "vn_STNC_reg_low", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "fc8e71c19cae2e6abfc731d8e046a01d52d4892d0497d8a83ec571b37e6f28c9", "task_id": "vn_STNC_reg_low-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0299019813537598, "target_z": -0.4003906732570357, "ce": 6.252951145172119, "genuineness": 4.444586753845215, "identity": 0.49767690896987915, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/de/a64bccd3d3bbb7517a5b8f95.mp3" }, { "key": "original-vn_STNC_reg_low-de-0-seed1701-scale0.5", "bucket": "vn_STNC_reg_low", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "64270d1b64de1fb349615ec26a15fe0adfb8423def040c00f8152391afa51317", "task_id": "vn_STNC_reg_low-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.729671001434326, "target_z": -0.1367186796592361, "ce": 6.133810997009277, "genuineness": 4.419482707977295, "identity": 0.5059568881988525, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/de/956b206afdfb7f3328786d50.mp3" }, { "key": "original-vn_STNC_reg_low-de-0-seed1701-scale0.75", "bucket": "vn_STNC_reg_low", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "422510ff286a3107eac180ca4590b84881fd9a1e859d9e9f0d9c6945db8f7f34", "task_id": "vn_STNC_reg_low-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6159725189208984, "target_z": -0.03686520830393199, "ce": 6.09409761428833, "genuineness": 4.482243061065674, "identity": 0.6880784034729004, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/de/1f0e2ce8d0e593910eee0cf5.mp3" }, { "key": "original-vn_STNC_reg_low-de-0-seed1701-scale1.0", "bucket": "vn_STNC_reg_low", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "668e7989f09f97e5b3347f0d5451c5539695626a2332981a9e8f457814921e5c", "task_id": "vn_STNC_reg_low-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.45890736579895, "target_z": 0.10107419441700319, "ce": 6.268836498260498, "genuineness": 4.557555198669434, "identity": 0.6075729727745056, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/de/2bea159f91b7b24bf38069ea.mp3" }, { "key": "original-vn_STNC_reg_low-de-0-seed1702-scale0.0", "bucket": "vn_STNC_reg_low", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "fa536606a42379258db4d8113fb5284e1af604ae9302487f62a04305bc4c2d81", "task_id": "vn_STNC_reg_low-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.39274525642395, "target_z": 0.159179774608039, "ce": 6.197352409362793, "genuineness": 4.6956281661987305, "identity": 0.46318989992141724, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/de/9c7eb53cae67b7a69695f1e6.mp3" }, { "key": "original-vn_STNC_reg_low-de-0-seed1702-scale0.5", "bucket": "vn_STNC_reg_low", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "3cb09db9a04c512861ba5ff81c559129060d7b0e543315ae512439eb9656a91b", "task_id": "vn_STNC_reg_low-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6293160915374756, "target_z": -0.04858394030978308, "ce": 6.173524379730225, "genuineness": 4.6203155517578125, "identity": 0.553551197052002, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/de/98ad30fb31998d85da9aabe5.mp3" }, { "key": "original-vn_STNC_reg_low-de-0-seed1702-scale0.75", "bucket": "vn_STNC_reg_low", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "4c4e2c5ea39955f9b6ba56cd00e67d26df796dd6bf8eba6ba37b2832a9270858", "task_id": "vn_STNC_reg_low-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8045153617858887, "target_z": 0.6757811894742264, "ce": 6.1020402908325195, "genuineness": 4.7834930419921875, "identity": 0.393324613571167, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/de/f0764fcd5d74edaf0c3f7795.mp3" }, { "key": "original-vn_STNC_reg_low-de-0-seed1702-scale1.0", "bucket": "vn_STNC_reg_low", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "fa163ae3ad88dd688bf11729337b0182473995e24a5c2f4abf73cc59b0650620", "task_id": "vn_STNC_reg_low-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3616104125976562, "target_z": 0.18652334303071824, "ce": 6.221180438995361, "genuineness": 4.758388519287109, "identity": 0.49167242646217346, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/de/fd28525055c5cea276a2dfa8.mp3" }, { "key": "original-vn_STNC_reg_low-de-1-seed1701-scale0.0", "bucket": "vn_STNC_reg_low", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "ff2ebab8e2a51d96b9b214c0ee8d760d440a53bdfd297f99d04f8bd5c15579f9", "task_id": "vn_STNC_reg_low-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5584282875061035, "target_z": 0.013671888904569609, "ce": 6.403861999511719, "genuineness": 4.545003414154053, "identity": 0.656863272190094, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/de/ab977f49afd301a3a6c0e19a.mp3" }, { "key": "original-vn_STNC_reg_low-de-1-seed1701-scale0.5", "bucket": "vn_STNC_reg_low", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "14eca5f38ebdb60ad6ecb94e5dcb2d76f4a7ea7cf33aae6e9f8feee8c305489f", "task_id": "vn_STNC_reg_low-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.061037063598633, "target_z": -0.4277344510661749, "ce": 6.427689552307129, "genuineness": 4.432034492492676, "identity": 0.4676640033721924, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/de/60a7fdc98512bc56ddc7fa9f.mp3" }, { "key": "original-vn_STNC_reg_low-de-1-seed1701-scale0.75", "bucket": "vn_STNC_reg_low", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "2b7bf05003f4574c3385069076c6f8c025264d466adb4f701f70dba9cf72868b", "task_id": "vn_STNC_reg_low-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.112187385559082, "target_z": -0.4726562221908608, "ce": 6.41180419921875, "genuineness": 4.319066047668457, "identity": 0.4651862680912018, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/de/3c1731a0fe43a5ad082a1526.mp3" }, { "key": "original-vn_STNC_reg_low-de-1-seed1701-scale1.0", "bucket": "vn_STNC_reg_low", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "3677552320c31c4b1ab010c84a13485ace507719934afa8f38b42eb291ee09df", "task_id": "vn_STNC_reg_low-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2011446952819824, "target_z": -0.5507812418208414, "ce": 6.3641486167907715, "genuineness": 4.319066047668457, "identity": 0.5754310488700867, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/de/caf70d157634a4eba27c1cd1.mp3" }, { "key": "original-vn_STNC_reg_low-de-1-seed1702-scale0.0", "bucket": "vn_STNC_reg_low", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "7cb3fb2d708916c98d91feb0492b90d657ac5e697761389aa195c578243ce939", "task_id": "vn_STNC_reg_low-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0855002403259277, "target_z": -0.4492187581791586, "ce": 6.300607204437256, "genuineness": 4.231201171875, "identity": 0.28974515199661255, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/de/62ca7341610d88ca23188c6c.mp3" }, { "key": "original-vn_STNC_reg_low-de-1-seed1702-scale0.5", "bucket": "vn_STNC_reg_low", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "1ad55ddc197f720f75fafba3d14e99dc49ce3886c22bf32270fed272b8f4b1cb", "task_id": "vn_STNC_reg_low-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.067708730697632, "target_z": -0.43359371237587047, "ce": 6.41180419921875, "genuineness": 4.507347106933594, "identity": 0.49850529432296753, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/de/db45759370268550594830ac.mp3" }, { "key": "original-vn_STNC_reg_low-de-1-seed1702-scale0.75", "bucket": "vn_STNC_reg_low", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "01c35729c84cdb36e315a5e6156231b17b2a39240a7f1806cfa104654ed1a42d", "task_id": "vn_STNC_reg_low-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.061037063598633, "target_z": -0.4277344510661749, "ce": 6.427689552307129, "genuineness": 4.632867813110352, "identity": 0.5089752674102783, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/de/d64f3403e1fb92f79e8c0e3d.mp3" }, { "key": "original-vn_STNC_reg_low-de-1-seed1702-scale1.0", "bucket": "vn_STNC_reg_low", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "95a7e1e3a0ba6efe64c219b3272ec6a76c0398b71d89ff026ff071f8cd4d9eb3", "task_id": "vn_STNC_reg_low-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8497633934020996, "target_z": -0.24218747709835595, "ce": 6.372091293334961, "genuineness": 4.821148872375488, "identity": 0.5882318019866943, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/de/598646e4d1304d978ab45d89.mp3" }, { "key": "original-vn_STNC_reg_low-en-0-seed1701-scale0.0", "bucket": "vn_STNC_reg_low", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "35d40365693de50ca63a776049d98216e99648221f8925ce7e81c3fc0c1e3dcc", "task_id": "vn_STNC_reg_low-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": 3.2500712871551514, "target_z": -0.5937500654332687, "ce": 6.157639026641846, "genuineness": 4.017816066741943, "identity": 0.38592612743377686, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/en/886ddf8eaa01842b7d3a668f.mp3" }, { "key": "original-vn_STNC_reg_low-en-0-seed1701-scale0.5", "bucket": "vn_STNC_reg_low", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "90300729da4ca6e0e9796cf845c3015543a32e13012c8b56c17ba3d22d59d4f8", "task_id": "vn_STNC_reg_low-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9824298620224, "target_z": 0.5195312549074952, "ce": 6.252951145172119, "genuineness": 4.256305694580078, "identity": 0.35365191102027893, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/en/42646c84ea21f4bd5317fa45.mp3" }, { "key": "original-vn_STNC_reg_low-en-0-seed1701-scale0.75", "bucket": "vn_STNC_reg_low", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "5dd7037dd57928e104a53fef0ab9b78a7c9ffce525f31469b4752f9f1844366d", "task_id": "vn_STNC_reg_low-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": 3.0143344402313232, "target_z": -0.3867187843524661, "ce": 6.117925643920898, "genuineness": 3.9676074981689453, "identity": 0.22110716998577118, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/en/c1b3699a5a687c7e2dbb97e0.mp3" }, { "key": "original-vn_STNC_reg_low-en-0-seed1701-scale1.0", "bucket": "vn_STNC_reg_low", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "ecdf095fd2463b688d309cef71b198dda00e6def046e471cd529b04f7d1522dc", "task_id": "vn_STNC_reg_low-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": 3.018782138824463, "target_z": -0.39062488876344315, "ce": 6.141753673553467, "genuineness": 4.061748504638672, "identity": 0.37383902072906494, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/en/8bb9079e78efb707f474553c.mp3" }, { "key": "original-vn_STNC_reg_low-en-0-seed1702-scale0.0", "bucket": "vn_STNC_reg_low", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "84b6cb204b9ce16502694ec5b4c1e2d75a880d88b7cfee60811487ba10bca3d5", "task_id": "vn_STNC_reg_low-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.511725664138794, "target_z": 0.054687555618278434, "ce": 6.23706579208374, "genuineness": 4.469690799713135, "identity": 0.22148117423057556, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/en/e97887706cf69a4c7852b214.mp3" }, { "key": "original-vn_STNC_reg_low-en-0-seed1702-scale0.5", "bucket": "vn_STNC_reg_low", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "1d60a43498ed75e24c7ae6ef6f71c42aefb6936691ca85f460269a5db955fc02", "task_id": "vn_STNC_reg_low-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": 3.067708730697632, "target_z": -0.43359371237587047, "ce": 6.157639026641846, "genuineness": 4.1998209953308105, "identity": 0.25791218876838684, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/en/71e5786d83e5077836881986.mp3" }, { "key": "original-vn_STNC_reg_low-en-0-seed1702-scale0.75", "bucket": "vn_STNC_reg_low", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "c2ef7d48c6b3f873355f7bb02ae3b3e44db15f74419fcb08f41fac2883ef7698", "task_id": "vn_STNC_reg_low-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.678520679473877, "target_z": -0.09179690853455023, "ce": 6.141753673553467, "genuineness": 4.3127899169921875, "identity": 0.19950339198112488, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/en/89f16590c9fb11bc7e570b3d.mp3" }, { "key": "original-vn_STNC_reg_low-en-0-seed1702-scale1.0", "bucket": "vn_STNC_reg_low", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "437b52dc67a538e2a1def95aaad3746c83f14ba74a2e16d9cfe1445e0b378329", "task_id": "vn_STNC_reg_low-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": 3.2945499420166016, "target_z": -0.6328125752482591, "ce": 6.292664527893066, "genuineness": 3.911123275756836, "identity": 0.2657482326030731, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/en/7bfe4edbd65fc558904425f5.mp3" }, { "key": "original-vn_STNC_reg_low-en-1-seed1701-scale0.0", "bucket": "vn_STNC_reg_low", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "a90e54666f9d89deb82551a1357cb371cc1c23741d9dc0866a792fda1db7135c", "task_id": "vn_STNC_reg_low-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.435555934906006, "target_z": 0.12158213246708759, "ce": 6.2767791748046875, "genuineness": 4.268857479095459, "identity": 0.4457080364227295, "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/vn_STNC_reg_low/en/dc8c9c37328b60dfc0d0c5d1.mp3" }, { "key": "original-vn_STNC_reg_low-en-1-seed1701-scale0.5", "bucket": "vn_STNC_reg_low", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "8d9c28bebd04724bf9690662d336d8ba6dd807fe885fcce415f700e0d396bd84", "task_id": "vn_STNC_reg_low-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.160344362258911, "target_z": 0.3632813203407639, "ce": 6.356205940246582, "genuineness": 4.645420074462891, "identity": 0.557519793510437, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/en/dc81ebf4396b59988ddc92ab.mp3" }, { "key": "original-vn_STNC_reg_low-en-1-seed1701-scale0.75", "bucket": "vn_STNC_reg_low", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "13ec061e6df55bc0b6ce7b77122bd37b4b7fc26323956ddd6e14c2c9086f73ee", "task_id": "vn_STNC_reg_low-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9023683071136475, "target_z": 0.5898437516358317, "ce": 6.316492557525635, "genuineness": 4.607763767242432, "identity": 0.29380807280540466, "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/vn_STNC_reg_low/en/14126dac4ab79bbc2dc243e4.mp3" }, { "key": "original-vn_STNC_reg_low-en-1-seed1701-scale1.0", "bucket": "vn_STNC_reg_low", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "b6e46a079f1cb0cdbdde5fbb091423d6be99669c33a12ad11e9d4cc894b615b8", "task_id": "vn_STNC_reg_low-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1492247581481934, "target_z": 0.3730468954478965, "ce": 6.300607204437256, "genuineness": 4.482243061065674, "identity": 0.3994259238243103, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/en/cc634ae1eff3493b8a1bc8af.mp3" }, { "key": "original-vn_STNC_reg_low-en-1-seed1702-scale0.0", "bucket": "vn_STNC_reg_low", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "7984e20b74c75d163ea42f58b8cb0562d4830a811a2db5a2be48bd1a6248fc90", "task_id": "vn_STNC_reg_low-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6168065071105957, "target_z": -0.037597642140951425, "ce": 6.332377910614014, "genuineness": 4.482243061065674, "identity": 0.41829490661621094, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/en/e96c05f7d3908d28438ad9ca.mp3" }, { "key": "original-vn_STNC_reg_low-en-1-seed1702-scale0.5", "bucket": "vn_STNC_reg_low", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "b074afe1cb580c4300236ffcc04687f6d9a8dcea71fd885cbc0fc4feb2f2148e", "task_id": "vn_STNC_reg_low-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7730376720428467, "target_z": -0.17480461102486716, "ce": 6.268836498260498, "genuineness": 4.406930446624756, "identity": 0.26647889614105225, "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/vn_STNC_reg_low/en/306a43955abd04faef567d2d.mp3" }, { "key": "original-vn_STNC_reg_low-en-1-seed1702-scale0.75", "bucket": "vn_STNC_reg_low", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "6b4f8de01e0368771e06ea69540b41c0fab5cc196f21046c99dbb56ca2533f83", "task_id": "vn_STNC_reg_low-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1492247581481934, "target_z": 0.3730468954478965, "ce": 6.3641486167907715, "genuineness": 4.507347106933594, "identity": 0.45257139205932617, "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/vn_STNC_reg_low/en/b31db9eea9d18af80e4aba2c.mp3" }, { "key": "original-vn_STNC_reg_low-en-1-seed1702-scale1.0", "bucket": "vn_STNC_reg_low", "language": "en", "sentences": [ "I found the little box.", "It was beside the 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "b9fb4524b61db6ceaffc0d6330875ba419a3326b364695e831d2cab111f21b4d", "task_id": "vn_STNC_reg_low-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3705060482025146, "target_z": 0.1787109248223042, "ce": 6.332377910614014, "genuineness": 4.532451152801514, "identity": 0.40078169107437134, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/en/a1ec64c5dad7d7d33141e51c.mp3" }, { "key": "original-vn_STNC_reg_low-es-0-seed1701-scale0.0", "bucket": "vn_STNC_reg_low", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "2b023761c43dba40ab4494de297ae34ef14dc7ce1ae3be5654dacf396c2dc123", "task_id": "vn_STNC_reg_low-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1700096130371094, "target_z": -0.5234374640117022, "ce": 6.0146708488464355, "genuineness": 4.18099308013916, "identity": 0.5916562080383301, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/es/3684d0efd9990a9a429f9456.mp3" }, { "key": "original-vn_STNC_reg_low-es-0-seed1701-scale0.5", "bucket": "vn_STNC_reg_low", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "03d3d1f0795da741246660e221087fc147ff5512835e190918f171a5d74d28b3", "task_id": "vn_STNC_reg_low-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.254518985748291, "target_z": -0.5976561698442457, "ce": 5.911416053771973, "genuineness": 4.250029563903809, "identity": 0.4589320421218872, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/es/e43430175814f208ce184207.mp3" }, { "key": "original-vn_STNC_reg_low-es-0-seed1701-scale0.75", "bucket": "vn_STNC_reg_low", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "9644a2bc22e885823a2a453048303375fc4113e5adcc4cc254863691e97595d8", "task_id": "vn_STNC_reg_low-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9298250675201416, "target_z": -0.3125000785199225, "ce": 6.149696350097656, "genuineness": 4.068024158477783, "identity": 0.731757402420044, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/es/44497ebef9221f5ffe4f9a12.mp3" }, { "key": "original-vn_STNC_reg_low-es-0-seed1701-scale1.0", "bucket": "vn_STNC_reg_low", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "4b3908d111aaadde4d4742e5755f3a6a610da39154d46b6d71908d4f5e037032", "task_id": "vn_STNC_reg_low-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9231531620025635, "target_z": -0.30664060782376695, "ce": 6.086154937744141, "genuineness": 4.306513786315918, "identity": 0.6528606414794922, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/es/f45eee45397533bde92f5f75.mp3" }, { "key": "original-vn_STNC_reg_low-es-0-seed1702-scale0.0", "bucket": "vn_STNC_reg_low", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "d016dfe7acc2d015aa008a1eb82c40e0535919b4e09ca1c7d63f4d09e5352d42", "task_id": "vn_STNC_reg_low-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5895633697509766, "target_z": -0.013671888904569609, "ce": 5.982900142669678, "genuineness": 4.457139015197754, "identity": 0.4706384241580963, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/es/2ebd372f879cd5b5860f1b91.mp3" }, { "key": "original-vn_STNC_reg_low-es-0-seed1702-scale0.5", "bucket": "vn_STNC_reg_low", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "dc2ba57dd556c801007dac1c0aef63bbd3917925739872066e995ac375a3fcf3", "task_id": "vn_STNC_reg_low-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.254518985748291, "target_z": -0.5976561698442457, "ce": 6.046441555023193, "genuineness": 3.8671910762786865, "identity": 0.6259295344352722, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/es/e3e62d9f5a6918b836ece382.mp3" }, { "key": "original-vn_STNC_reg_low-es-0-seed1702-scale0.75", "bucket": "vn_STNC_reg_low", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "f5f304a309d1ff9ea8c68a507835d49b4e7a8d2c02b4da5944fd5b4fa898223f", "task_id": "vn_STNC_reg_low-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.681856632232666, "target_z": -0.094726643882628, "ce": 6.054384231567383, "genuineness": 4.469690799713135, "identity": 0.4616713225841522, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/es/722a9d396a29d39b4ec82bec.mp3" }, { "key": "original-vn_STNC_reg_low-es-0-seed1702-scale1.0", "bucket": "vn_STNC_reg_low", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "2292f93b7d412c9bd7c44ec86ac27cd0687641eeae63b5fc2a4c7120ebc5610e", "task_id": "vn_STNC_reg_low-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4605753421783447, "target_z": 0.0996093267429643, "ce": 6.062326908111572, "genuineness": 4.444586753845215, "identity": 0.5777477025985718, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/es/47a75e974244671966be05ce.mp3" }, { "key": "original-vn_STNC_reg_low-es-1-seed1701-scale0.0", "bucket": "vn_STNC_reg_low", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "e1f264eb6290c3bea73bbce440de9a1e3ae2bfbcd311f921e0b173824e990bd4", "task_id": "vn_STNC_reg_low-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(submissive and deferential. 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(submissive and deferential. 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": 1.8178589344024658, "target_z": 0.6640624574683753, "ce": 6.1894097328186035, "genuineness": 4.7081804275512695, "identity": 0.593679666519165, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/es/eb5d351709093fe2cb4dc76e.mp3" }, { "key": "original-vn_STNC_reg_low-es-1-seed1701-scale0.5", "bucket": "vn_STNC_reg_low", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "341ba806e8f2d852f0932679bf584559899624802a0b9b5bf910f7c1a66b2bb7", "task_id": "vn_STNC_reg_low-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9298250675201416, "target_z": -0.3125000785199225, "ce": 6.221180438995361, "genuineness": 4.582659721374512, "identity": 0.5878376960754395, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana. Vieja, podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/es/6beb2d952d785035c363ff75.mp3" }, { "key": "original-vn_STNC_reg_low-es-1-seed1701-scale0.75", "bucket": "vn_STNC_reg_low", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "8925662b038c9da656a5e38dea1c18ad433788028bc2dd74f8337dd15a77b25c", "task_id": "vn_STNC_reg_low-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3182435035705566, "target_z": 0.2246094837828093, "ce": 6.109982967376709, "genuineness": 4.6705241203308105, "identity": 0.6433123350143433, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/es/646c0d793130ff73c0eaaa89.mp3" }, { "key": "original-vn_STNC_reg_low-es-1-seed1701-scale1.0", "bucket": "vn_STNC_reg_low", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "5705993f7279e3ba34f18c8ceeda28ebb712d611006688d86118968d38ea52c4", "task_id": "vn_STNC_reg_low-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.914257526397705, "target_z": -0.2988281896153529, "ce": 6.165581703186035, "genuineness": 4.432034492492676, "identity": 0.41797417402267456, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/es/bdd81d507b3079873542f2ed.mp3" }, { "key": "original-vn_STNC_reg_low-es-1-seed1702-scale0.0", "bucket": "vn_STNC_reg_low", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "43335921a52f9ec36cc9989409fd7350c6c0aabf206e4a63b5de8a8017a4d5bb", "task_id": "vn_STNC_reg_low-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4605753421783447, "target_z": 0.0996093267429643, "ce": 6.165581703186035, "genuineness": 4.7081804275512695, "identity": 0.4569689631462097, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/es/c64be6024ce575e02c295d85.mp3" }, { "key": "original-vn_STNC_reg_low-es-1-seed1702-scale0.5", "bucket": "vn_STNC_reg_low", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "bc824dbb9a43368c33efa2ed89a788267f8c9264e597eb5b5448e53f77d2a6f8", "task_id": "vn_STNC_reg_low-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.761918067932129, "target_z": -0.16503903591773458, "ce": 6.109982967376709, "genuineness": 4.482243061065674, "identity": 0.7166200876235962, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay.", "wer": 0.058823529411764705, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/es/d5f4dba247a8fa224cedef40.mp3" }, { "key": "original-vn_STNC_reg_low-es-1-seed1702-scale0.75", "bucket": "vn_STNC_reg_low", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "2323592fba9f9f1cac8fd00f218f2b17c21a9b6b351af7ecd04d6474e9823e38", "task_id": "vn_STNC_reg_low-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5400807857513428, "target_z": 0.02978522393253736, "ce": 6.125868320465088, "genuineness": 4.7834930419921875, "identity": 0.8437378406524658, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/es/126b79555592385e98331d14.mp3" }, { "key": "original-vn_STNC_reg_low-es-1-seed1702-scale1.0", "bucket": "vn_STNC_reg_low", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "7eadc653fcececff8b04e23a3b883bcc2e78001cb89d37612566de24e707b4b5", "task_id": "vn_STNC_reg_low-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5000500679016113, "target_z": 0.06494141995009065, "ce": 6.24500846862793, "genuineness": 4.65797233581543, "identity": 0.5366983413696289, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/es/5312da1c4bf64e593f54010d.mp3" }, { "key": "original-vn_STNC_reg_low-fr-0-seed1701-scale0.0", "bucket": "vn_STNC_reg_low", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "b4ee00312502c4fc411bc08f131a336ada5c3b51eb426acaebf8fa8875f7b73f", "task_id": "vn_STNC_reg_low-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": 1.8445460796356201, "target_z": 0.6406249934566731, "ce": 6.125868320465088, "genuineness": 4.770940780639648, "identity": 0.8228175044059753, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/fr/4cd43688e3e9737f059f9dc9.mp3" }, { "key": "original-vn_STNC_reg_low-fr-0-seed1701-scale0.5", "bucket": "vn_STNC_reg_low", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "8735d7fc6f0b2b03d03d500937e03c40f269166626952785bcd86a7a64cc0c89", "task_id": "vn_STNC_reg_low-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": 1.9068162441253662, "target_z": 0.5859374378383947, "ce": 6.221180438995361, "genuineness": 4.7332844734191895, "identity": 0.7977349162101746, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/fr/94d90e6b738bde1aec14177a.mp3" }, { "key": "original-vn_STNC_reg_low-fr-0-seed1701-scale0.75", "bucket": "vn_STNC_reg_low", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "f406c899c023ba7053a59aba669966125ff769424e8e8e4fb4e82fb07a78dd14", "task_id": "vn_STNC_reg_low-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8330838680267334, "target_z": -0.22753900974442706, "ce": 6.133810997009277, "genuineness": 4.432034492492676, "identity": 0.7922347784042358, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/fr/91c7d3b29262be4e36b90a27.mp3" }, { "key": "original-vn_STNC_reg_low-fr-0-seed1701-scale1.0", "bucket": "vn_STNC_reg_low", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "bb00b5f3172dd6eeb6a20bac97365d196479e9fd981166ac10ad8e263eeb0976", "task_id": "vn_STNC_reg_low-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2604212760925293, "target_z": 0.2753907256036507, "ce": 6.0146708488464355, "genuineness": 4.645420074462891, "identity": 0.8102103471755981, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/fr/479f3c1e50b2bdc82d3deb60.mp3" }, { "key": "original-vn_STNC_reg_low-fr-0-seed1702-scale0.0", "bucket": "vn_STNC_reg_low", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "f01aa1f65293fa07eb193ae7960d1e341e2c58e60d63e2b1b40798176d8533ef", "task_id": "vn_STNC_reg_low-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": 1.7600367069244385, "target_z": 0.7148436992892168, "ce": 6.141753673553467, "genuineness": 4.871357440948486, "identity": 0.7797678709030151, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/fr/03dd8650e63c6af0ff1caf56.mp3" }, { "key": "original-vn_STNC_reg_low-fr-0-seed1702-scale0.5", "bucket": "vn_STNC_reg_low", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "56bf047630bd6fa34c8eaefb8a95a683978fd386743057ce94d03a7a093f0dce", "task_id": "vn_STNC_reg_low-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6173624992370605, "target_z": -0.03808593136563106, "ce": 6.149696350097656, "genuineness": 4.457139015197754, "identity": 0.738010048866272, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/fr/92a368ae6bbc2280597e5b01.mp3" }, { "key": "original-vn_STNC_reg_low-fr-0-seed1702-scale0.75", "bucket": "vn_STNC_reg_low", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "c149d057501d3a9ca379443177e7b0102fee478998b994baf6c1bd555bc341fb", "task_id": "vn_STNC_reg_low-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2848846912384033, "target_z": 0.253906209104207, "ce": 6.070269584655762, "genuineness": 4.632867813110352, "identity": 0.7953929901123047, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/fr/44cc7f0b741ada3c0da33d7f.mp3" }, { "key": "original-vn_STNC_reg_low-fr-0-seed1702-scale1.0", "bucket": "vn_STNC_reg_low", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "36a98cda692876a766c68d4e253a381ecbe1edc0a801e81f9359789e851b92ca", "task_id": "vn_STNC_reg_low-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6484975814819336, "target_z": -0.06542970917477027, "ce": 6.038498878479004, "genuineness": 4.319066047668457, "identity": 0.5753674507141113, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/fr/49788297ce26833e1c555aa0.mp3" }, { "key": "original-vn_STNC_reg_low-fr-1-seed1701-scale0.0", "bucket": "vn_STNC_reg_low", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "933a0ffbfa232c00c20ca4535ec0a61b385c0f3d025fad1fa0abd6ea7ea9d299", "task_id": "vn_STNC_reg_low-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.07361102104187, "target_z": 0.439453183072026, "ce": 6.221180438995361, "genuineness": 4.7332844734191895, "identity": 0.7878793478012085, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/fr/3aeedafaf58f1aa6227c1d2c.mp3" }, { "key": "original-vn_STNC_reg_low-fr-1-seed1701-scale0.5", "bucket": "vn_STNC_reg_low", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "6b28144f8c9d59d845357e7beddae5f01c7e470721248218276cd7e4c466505e", "task_id": "vn_STNC_reg_low-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6284821033477783, "target_z": -0.047851506472763636, "ce": 6.1020402908325195, "genuineness": 4.570107460021973, "identity": 0.513563871383667, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/fr/4f134482b1956db097ec80b8.mp3" }, { "key": "original-vn_STNC_reg_low-fr-1-seed1701-scale0.75", "bucket": "vn_STNC_reg_low", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "f031eae8167763cb58a67d5ca8980c2ee74e005d3c956fcc53eaf9ce044fddd3", "task_id": "vn_STNC_reg_low-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9824298620224, "target_z": 0.5195312549074952, "ce": 6.046441555023193, "genuineness": 4.808597087860107, "identity": 0.8359663486480713, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/fr/d9eb7aad036a3deaa5527dec.mp3" }, { "key": "original-vn_STNC_reg_low-fr-1-seed1701-scale1.0", "bucket": "vn_STNC_reg_low", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "6267100f8264c64ad778880711901e1756a1b70bb46ecd2dc3f3bdbcd67f6679", "task_id": "vn_STNC_reg_low-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4689149856567383, "target_z": 0.09228519775922986, "ce": 6.125868320465088, "genuineness": 4.7081804275512695, "identity": 0.9097310900688171, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/fr/e0f31973d2652e2804b4744b.mp3" }, { "key": "original-vn_STNC_reg_low-fr-1-seed1702-scale0.0", "bucket": "vn_STNC_reg_low", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "40f450c6b1a682a7b23d6fb71387e8dea311b526597ed11ade536b95dbd62ca7", "task_id": "vn_STNC_reg_low-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5684359073638916, "target_z": 0.004882892246796283, "ce": 6.09409761428833, "genuineness": 4.413206577301025, "identity": 0.8123661279678345, "hypothesis": "J'ai trouvé la petite boîte elle était près de la vieille fenêtre nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/fr/75f2008586eb85cfb40a8fb8.mp3" }, { "key": "original-vn_STNC_reg_low-fr-1-seed1702-scale0.5", "bucket": "vn_STNC_reg_low", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "854bb659737d7e3e5bd2e1b4274495d3b2aacd32a17e23d5e6b938e1a71dedb5", "task_id": "vn_STNC_reg_low-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.227062463760376, "target_z": 0.30468745092504845, "ce": 6.173524379730225, "genuineness": 4.796044826507568, "identity": 0.8207789659500122, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/fr/154f272b17a67be1eee743ee.mp3" }, { "key": "original-vn_STNC_reg_low-fr-1-seed1702-scale0.75", "bucket": "vn_STNC_reg_low", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "17ad9d8f76be99029e954d9b3df30772f30baebb34bc9c6222e4ca1590ed0ce9", "task_id": "vn_STNC_reg_low-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3149077892303467, "target_z": 0.22753900974442706, "ce": 6.078212261199951, "genuineness": 4.6956281661987305, "identity": 0.7949323654174805, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/fr/f47370fe86ce77a6694d4ea3.mp3" }, { "key": "original-vn_STNC_reg_low-fr-1-seed1702-scale1.0", "bucket": "vn_STNC_reg_low", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "733a7a851d23935b8772be217c2b51e7b3f08d8da5c69f657aaf9a865a5af771", "task_id": "vn_STNC_reg_low-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4327762126922607, "target_z": 0.12402336920402575, "ce": 6.046441555023193, "genuineness": 4.582659721374512, "identity": 0.7432072758674622, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/fr/03c92c77a1c4bdb58511256a.mp3" }, { "key": "original-vn_STNC_reg_low-nl-0-seed1701-scale0.0", "bucket": "vn_STNC_reg_low", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "6438a4402218e620095c669a65237cf91eb94a5f7950bfc9886daafcb9d48adf", "task_id": "vn_STNC_reg_low-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4294402599334717, "target_z": 0.1269531045521035, "ce": 6.109982967376709, "genuineness": 4.469690799713135, "identity": 0.7786049246788025, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/nl/967526dcbf9bcae11897aff5.mp3" }, { "key": "original-vn_STNC_reg_low-nl-0-seed1701-scale0.5", "bucket": "vn_STNC_reg_low", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "4365d23adca96f00f47cc3d2b1022043ac48e1dda2a916b516ab594e462631a5", "task_id": "vn_STNC_reg_low-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.570659875869751, "target_z": 0.0029297353480777695, "ce": 5.974957466125488, "genuineness": 4.356721878051758, "identity": 0.6068152785301208, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/nl/7070e6a74de80727c2063b20.mp3" }, { "key": "original-vn_STNC_reg_low-nl-0-seed1701-scale0.75", "bucket": "vn_STNC_reg_low", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "fcc82cfcfba846916662749b947510b02c73c249a3b2d63e7ea8f56999772d42", "task_id": "vn_STNC_reg_low-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1099634170532227, "target_z": -0.4707030652921423, "ce": 6.109982967376709, "genuineness": 4.306513786315918, "identity": 0.6751308441162109, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/nl/1e524e0298e35f96a85b111b.mp3" }, { "key": "original-vn_STNC_reg_low-nl-0-seed1701-scale1.0", "bucket": "vn_STNC_reg_low", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "5de0ba2b67ce86231d2201c1d800a2b4e064398b670f26ee766fcabeea542c38", "task_id": "vn_STNC_reg_low-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0276780128479004, "target_z": -0.3984375163583172, "ce": 6.157639026641846, "genuineness": 4.268857479095459, "identity": 0.6360951066017151, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/nl/9bc2ae00ee6f2333acefc925.mp3" }, { "key": "original-vn_STNC_reg_low-nl-0-seed1702-scale0.0", "bucket": "vn_STNC_reg_low", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "e8483de201b797d6ba50f545cc37286b9777f23f39b2b5b1e3db8b8876d5611d", "task_id": "vn_STNC_reg_low-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4680166244506836, "target_z": -0.7851563007107832, "ce": 6.109982967376709, "genuineness": 4.193545341491699, "identity": 0.7322454452514648, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/nl/0eeb2a43cf56d60dc7ef7787.mp3" }, { "key": "original-vn_STNC_reg_low-nl-0-seed1702-scale0.5", "bucket": "vn_STNC_reg_low", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "724c6d57c8cb4d77f328ef89b72fd078658fc5ff397ae14b5581d0e7d0ec0aa7", "task_id": "vn_STNC_reg_low-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6557252407073975, "target_z": -0.07177725970914546, "ce": 6.165581703186035, "genuineness": 4.482243061065674, "identity": 0.7036973237991333, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/nl/8cc336f9312834835b7d1797.mp3" }, { "key": "original-vn_STNC_reg_low-nl-0-seed1702-scale0.75", "bucket": "vn_STNC_reg_low", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "b62ae83b9a4027f1ec9b2d42efe301899bd8fb5f8e05243ebdd29e258be99700", "task_id": "vn_STNC_reg_low-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0157887935638428, "target_z": 0.4902344248928674, "ce": 5.967014789581299, "genuineness": 4.381826400756836, "identity": 0.6273688077926636, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/nl/81ce31b874aa9a176584e8d4.mp3" }, { "key": "original-vn_STNC_reg_low-nl-0-seed1702-scale1.0", "bucket": "vn_STNC_reg_low", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "0d751aeb7fedeaac8f2388c99ca0d0484490b2cea1fb5554843fd4a766603ff2", "task_id": "vn_STNC_reg_low-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3924028873443604, "target_z": -0.7187500130866538, "ce": 5.9431867599487305, "genuineness": 4.281409740447998, "identity": 0.666627049446106, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/nl/1283b8ce03888cb57f0a259f.mp3" }, { "key": "original-vn_STNC_reg_low-nl-1-seed1701-scale0.0", "bucket": "vn_STNC_reg_low", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "4c6d57053ce1f3089d95ab0c8a2d2ddfc9d6d312bc1eca4b813bdf8d23b0380d", "task_id": "vn_STNC_reg_low-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7363429069519043, "target_z": -0.14257815035539165, "ce": 6.09409761428833, "genuineness": 4.293961524963379, "identity": 0.6190439462661743, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/nl/69d075e33b912826092fb1e8.mp3" }, { "key": "original-vn_STNC_reg_low-nl-1-seed1701-scale0.5", "bucket": "vn_STNC_reg_low", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "4990b5207edda195bf6cc5444dfe0e4bd266b1384c30508232bafb49f6f52f4c", "task_id": "vn_STNC_reg_low-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.410536766052246, "target_z": 0.1435547288047509, "ce": 6.062326908111572, "genuineness": 4.432034492492676, "identity": 0.656312882900238, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/nl/564888f98c865c7c234e713e.mp3" }, { "key": "original-vn_STNC_reg_low-nl-1-seed1701-scale0.75", "bucket": "vn_STNC_reg_low", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "f620f5cea6069a5edc4069be8fdceb5982ab614a3cab2ac6df29435d30a4bf7d", "task_id": "vn_STNC_reg_low-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4190900325775146, "target_z": -0.742187477098356, "ce": 6.2767791748046875, "genuineness": 4.595211505889893, "identity": 0.4247932434082031, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/nl/e0546877059fbe39981fdd6e.mp3" }, { "key": "original-vn_STNC_reg_low-nl-1-seed1701-scale1.0", "bucket": "vn_STNC_reg_low", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "2567637aa494c6c5ccb3036b03d1b1a95bc77fe0ae170b7c7b83073675f98729", "task_id": "vn_STNC_reg_low-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.387185573577881, "target_z": 0.1640624574683753, "ce": 5.911416053771973, "genuineness": 4.444586753845215, "identity": 0.4752432703971863, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/nl/0334b5bbc707b8b11104b697.mp3" }, { "key": "original-vn_STNC_reg_low-nl-1-seed1702-scale0.0", "bucket": "vn_STNC_reg_low", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "c9da8aa49d24b90ff6395348caaec6aeb0532ee5125705e6fcee4b2bd2bc8088", "task_id": "vn_STNC_reg_low-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4900424480438232, "target_z": 0.07373041660786397, "ce": 5.974957466125488, "genuineness": 4.356721878051758, "identity": 0.7278918027877808, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/nl/00826184838a93cedc571290.mp3" }, { "key": "original-vn_STNC_reg_low-nl-1-seed1702-scale0.5", "bucket": "vn_STNC_reg_low", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "32daecf8fe06ab4d291172225a5ed88ad57a3fcf2154d2930565ddce74ebd3be", "task_id": "vn_STNC_reg_low-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5425827503204346, "target_z": 0.02758792242147903, "ce": 6.117925643920898, "genuineness": 4.444586753845215, "identity": 0.5676547288894653, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/nl/ba324aeb436fa6c9b450700f.mp3" }, { "key": "original-vn_STNC_reg_low-nl-1-seed1702-scale0.75", "bucket": "vn_STNC_reg_low", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "d7c3b36ffe3002226986f01a7d805d57d573f368cb3531b1c886f822691dca5d", "task_id": "vn_STNC_reg_low-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.066939353942871, "target_z": 0.4453124443817216, "ce": 6.078212261199951, "genuineness": 4.293961524963379, "identity": 0.6998186111450195, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/nl/6f9df71878a492d0aeebf5a0.mp3" }, { "key": "original-vn_STNC_reg_low-nl-1-seed1702-scale1.0", "bucket": "vn_STNC_reg_low", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We 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. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. 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": [ "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 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": "33ea3c69ac71b3c80b1306d4b1a2479c8a8552028656102bf2aa95a5ab66635a", "mp3_sha256": "925697070a4c56b200ae383e1c03900edbd3a70d8426a98e08ee1d0642eead4d", "task_id": "vn_STNC_reg_low-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(submissive and deferential. 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(submissive and deferential. 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(submissive and deferential. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.657735824584961, "target_z": 0.8046874509250485, "ce": 5.974957466125488, "genuineness": 4.400654315948486, "identity": 0.5971636772155762, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_STNC_reg_low/nl/4572ecaa685c6bd7b9ff35f3.mp3" } ], "source_standard_audio_unchanged": true }